ArticlePDF Available

Autonomous Intelligent Vehicles (AIV): Research statements, open issues, challenges and road for future

January 2021
International Journal of Intelligent Networks 2(4):83-102

January 2021
2(4):83-102

DOI:10.1016/j.ijin.2021.07.002

License
CC BY-NC-ND 4.0

Authors:

Amit Tyagi

Pondicherry Engineering College

Dr aswathy Su

Jyothi Engineering College

The vehicle business has accomplished extraordinary outcomes in creating dependable, sheltered and reasonable vehicles throughout the only remaining century. Autonomous Vehicles (AV) are turning into a reality due to generous late improvements in Internet of Things (IoTs), Programmable Logic Controller (PLC) and correspondence innovations in computing filed. Notice that self-sufficient vehicles here are for vehicles just including transports, light vehicles, and so forth Model models have now voyaged a huge number of miles in test driving for self-ruling vehicles (under self-sufficient vehicles). This isn't a final attempt to self-governing vehicles; we have to plan Autonomous Intelligent Vehicles (AIV) that can make a compelling (for example, without hurting people) choice over the street organization. Later on, self-governing vehicles will be a reality, in any event, for commercialization, as by and by. In AIV, we face various specialized and non-specialized issues, for example, programming multifaceted nature, constant information examination and testing and confirmation. Handling these issues requires viable and prompt arrangements that meet the prerequisites, guidelines and strategies of clients, industry and government. The analysis of this work will help numerous research analysts who work in Autonomous Vehicles or Intelligent Transport Systems today and so on in near future to get better solution.

Smart applications in current scenarios.

…

Progress towards autonomous intelligent vehicles (from 2000 to 2020).

…

Key organization of our work.

…

High-Level Functional Parts of a standard Intelligent Autonomous Vehicle System.

…

Current and future challenges for Autonomous Intelligent Vehicle.

…

Figures - uploaded by Amit Tyagi

Content may be subject to copyright.

Content uploaded by Amit Tyagi

Content may be subject to copyright.

Autonomous Intelligent Vehicles (AIV): Research statements, open issues,

challenges and road for future

Amit Kumar Tyagi

, S U Aswathy

School of Computer Science and Engineering, Vellore Institute of Technology, Chennai, 600127, Tamil Nadu, India

Centre for Advanced Data Science, Vellore Institute of Technology, Chennai, 600127, Tamilnadu, India

Department of Computer Science and Engineering, Jyothi Engineering College, Cheruthuruthy, Thrissur, Kerala, India

ARTICLE INFO

Keywords:

Autonomous intelligent vehicles (AIV)

Radio detection and ranging (RADAR)

Privacy

Security

Intelligent transportation system (ITS)

Sensors

Laser and vehicle navigation

ABSTRACT

The vehicle business has accomplished extraordinary outcomes in creating dependable, sheltered and reasonable

vehicles throughout the only remaining century. Autonomous Vehicles (AV) are turning into a reality due to

generous late improvements in Internet of Things (IoTs), Programmable Logic Controller (PLC) and correspon-

dence innovations in computing ﬁled. Notice that self-sufﬁcient vehicles here are for vehicles just including

transports, light vehicles, and so forth Model models have now voyaged a huge number of miles in test driving for

self-ruling vehicles (under self-sufﬁcient vehicles). This isn't a ﬁnal attempt to self-governing vehicles; we have to

plan Autonomous Intelligent Vehicles (AIV) that can make a compelling (for example, without hurting people)

choice over the street organization. Later on, self-governing vehicles will be a reality, in any event, for

commercialization, as by and by. In AIV, we face various specialized and non-specialized issues, for example,

programming multifaceted nature, constant information examination and testing and conﬁrmation. Handling

these issues requires viable and prompt arrangements that meet the prerequisites, guidelines and strategies of

clients, industry and government. The analysis of this work will help numerous research analysts who work in

Autonomous Vehicles or Intelligent Transport Systems today and so on in near future to get better solution.

1. Introduction

Vehicles have made unmatched steps in the car business and data

innovation, changing the traditional car from a good old wellspring of

transport into a full-scale (moving/voyaging) gadget and driving

framework wealthy in infotainment. The presentation of very good

quality vehicles today gives the premise to the acknowledgment of clever

vehicles in numerous ﬁelds, for example, medical services, co-

ordination's, and so forth These keen vehicles are self-governing (driv-

erless) in that they underwrite attributes, for example, detecting the

world, settling on quick and opportune choices, exploring out and about

without human info, keeping up secure examples of portability, playing

out a wide range of moves, and to give some examples, journey control.

Autonomous Intelligent Vehicles (AIV) has been alluded to as such ve-

hicles. A self-ruling astute vehicle alludes to a PC controlled vehicle that,

with no human intercession, can guide itself, acclimate itself with the

climate, decide, and work totally [1–3]. Self-governing vehicles [1,2] are

fundamentally worried about: wiping out driver needs, inferable from

monstrous populace development, extending foundation, expanding the

quantity of vehicles, the requirement for fruitful time the executives, and

the usage and enhancement of assets. This stressfully affects our trans-

portation framework as the human populace extends and the quantity of

vehicles increments, going from streets and parking spots to fuel stations

(for vehicles with fuel motors) and charging stations (for electric and

cross breed vehicles). There are several smart applications exist in cur-

rent which can be depicted as Fig. 1.

In Fig. 1, we can ﬁnd out unmanned aerial vehicle are the example/

type of intelligent transportation systems.

Autonomous vehicles are considered as potential vehicles and

referred to as Vehicles of Tomorrow (or Internet of Vehicles (IoV)) as sub-

components of Autonomous Intelligent Vehicles. Remember that the

vehicles of tomorrow are autonomous vehicles, the Internet of Vehicles,

electric vehicles, hybrid electric vehicles. We have seen tremendous

improvements in transport or in the ﬁeld of vehicles in the previous

century (or years) (refer Fig. 2). In the transportation ﬁeld, we have seen

many signiﬁcant fatalities in the past decade. We are counting/having

more millions of road accidents around the world; many people are

losing their lives as a result. The number of deaths is rising year by year

* Corresponding author. School of Computer Science and Engineering, Vellore Institute of Technology, Chennai, 600127, Tamil Nadu, India.

E-mail addresses: amitkrtyagi025@gmail.com (A.K. Tyagi), aswathy.su@gmail.com (S.U. Aswathy).

Contents lists available at ScienceDirect

International Journal of Intelligent Networks

journal homepage: www.keaipublishing.com/en/journals/

international-journal-of-intelligent-networks

https://doi.org/10.1016/j.ijin.2021.07.002

Received 8 May 2021; Received in revised form 23 July 2021; Accepted 25 July 2021

license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

International Journal of Intelligent Networks 2 (2021) 83–102

due to the increasing number of vehicles. It was because of human error

that we found out the reason for these fatalities, i.e., not responding

instantly or to a critical situation or lack of knowledge, i.e. not having

adequate driving vehicle training. In many applications, human error is

the key issue, in which millions of lives are lost annually. Consequently,

in the present (brilliant) time, elective advances, for example, associated

vehicles and self-ruling vehicles should be investigated to limit human

mix-ups and decrease dangerous circumstances out and about. None-

theless, a few nations/governments are currently ﬁnding a way to make

the vehicle business lethal or mishap free, for example utilizing street

security, for example, Closed-Circuit Television (CCTV) cameras, trafﬁc

sensors, and so forth.

Amazing exploration discoveries from the ﬁelds of remote corre-

spondence, installed frameworks, route, sensor and adhoc network

innovation, information assortment and spread, and information inves-

tigation are the consequence of the advancement and development of

Autonomous Vehicles. During the 1920s, the idea of Autonomous Vehi-

cles started with "ghost vehicles" where the vehicle was constrained by a

controller gadget [4]. We saw the presentation of Autonomous Vehicles

during the 1980s that were independent and independent. The NavLab at

Carnegie Mellon University, where analysts built up the Autonomous

Land Vehicle (ALV) [5], was a signiﬁcant supporter of the self-governing

keen vehicle area. The Prometheus venture supported by Mercedes in

1987 [6], accomplished a major outcome in the very decade with the

plan of their ﬁrst automated vehicle to follow path markings and

different vehicles (regardless, human obstruction was needed for security

reasons). While it was not completely self-sufﬁcient at that point, a huge

headway was the capacity to move to another lane consequently. The

requirement for independent and smart) vehicles have now expanded in

the 21st century, basically because of minimal effort, elite advancements

in different regions. Notice that the qualiﬁcation between two differen-

tiating ideas, for example, the robotized vehicle and the self-ruling

insightful vehicle might be incorporated here as: self-sufﬁcient vehicle

alludes to a vehicle worked by a PC that may require human intercession

(for example a crisis brake, journey control, shrewd park, and so forth),

while independent clever vehicle centres around the activities

performed.

Connected vehicle innovation is utilized via Autonomous Vehicles

[7]. Note that speciﬁc advancements are shared among Autonomous

Intelligent Vehicles, Autonomous Vehicles/Intelligent Vehicles and

Connected Vehicles. For instance, the associated vehicle utilizes a

specially appointed vehicle organization (VANET) innovation where an

On-Board Device (OBU) is mounted on the vehicle; and vehicles can

speak with one another when they are inside their correspondence range

by means of the Dedicated Short-Range Communication (DSRC) standard

convention [8]. Two wide classes of utilizations are empowered by

Vehicular Adhoc Network (VANET) innovation: wellbeing related ap-

plications and data, amusement (all things considered called infotain-

ment) applications. Safety efforts for correspondence are exacting in

wellbeing related applications, while for data, diversion (all in all, info-

tainment) applications, safety efforts are moderately loose. VANET/fu-

ture vehicle innovation usage and enhancements, for example,

VANET-based mists or Vehicular Cloud Computing [9], Internet of

Things (IoT)- based Vehicles or Internet of Vehicles (IoVs) incorporate

mishap alert, crash notice, street development, trafﬁc lights, rescue

vehicle approach notice, over the top speed, dark ice on the asphalt, mist

cautioning, trafﬁc data, Internet development, trafﬁc lights, rescue

vehicle approach notice, inordinate speed, dark ice on the asphalt, mist

cautioning, trafﬁc data Notice that numerous VANET foundation appli-

cations/administrations rely upon agreeable contact among vehicles and

framework (Certiﬁed Authority) [10].

We anticipate that future vehicles should move towards the utiliza-

tion of Blockchain Technology sooner rather than later for message

transmission/secure correspondence. In that additionally, participation

Fig. 1. Smart applications in current scenarios.

Fig. 2. Progress towards autonomous intelligent vehicles (from 2000 to 2020).

A.K. Tyagi, S.U. Aswathy International Journal of Intelligent Networks 2 (2021) 83–102

and check of the square created in obligatory, Blockchain Technology

with this cycle (agreement) is a basic methodology, not a total veriﬁca-

tion arrangement against danger. As per the IEEE 802.11p norm, vehicles

share their area data with their neighbours, for example, position, sep-

aration, speeding up, and other control data. Between vehicular corre-

spondence additionally requires ﬁne-grained input from different

innovations, for example, exact situating frameworks, tangible data from

vehicles and solid and exact route. In this sense, world-driving vehicle

producers have been planning and coordinating a wide assortment of

new highlights into their very good quality vehicles (alongside scholarly

exploration endeavours). Brilliant parking, occurrence cautioning, crisis

slowing down, and self-loader and completely programmed (restricted)

pilot driving are the activities taken by driving vehicle makers, for

example, Audi, BMW, Toyota, Honda, Kia, Hyundai, Mercedes, Ford,

Nissan, Tesla, GM, Volvo, Bosch, and Volkswagen, all of which have

expanded rivalry in the associated vehicle/self-governing vehicle in-

dustry. What's more, ofﬁcials are dealing with enacting contact from

vehicle to vehicle (V2V) to permit normal customers to receive the re-

wards of innovation [11]. In any case, human activities and driving

propensities keep on assuming an essential part in safe driving, even with

the utilization of cutting-edge innovations.

Google, Audi and Tesla organizations have arisen as the biggest

market in earlier years for the creation of self-sufﬁcient, related car in-

novations and driverless vehicle innovation. Today, with the participa-

tion of a few organizations (innovation ﬁrms and vehicle makers)

cooperating to encourage the plan and creation of driverless vehicles, the

world is moving so rapidly. In earlier years for instance, Google had

shaped a partnership with Volvo and Toyota for the creation of driverless

vehicles [12]. Additionally, by dispatching its lead NVidia Drive PX2,

NVidia has likewise indicated a solid interest in Autonomous Vehicles.

Note that NVidia Drive PX2 is a solid Autonomous Vehicle registering

stage dependent on GPUs [13]. Uber and Apple, together, are addition-

ally arising partners in the self-sufﬁcient canny vehicle industry. Good-

bye, Yutong, KIA, and Hyundai are signiﬁcant organizations in the Asian

market that put resources into the plan, creation, and examination of

self-sufﬁcient vehicles. The objective for the European car industry is to

accomplish the effective acknowledgment by 2025 of self-governing

vehicles. Also, Mercedes and BMW are two other driving European or-

ganizations right now dealing with the idea of driverless model vehicles,

and sooner rather than later they intend to construct undeniable business

models. In the course of recent many years, vehicle proprietorship has

expanded dramatically as their costs drop; as their wages rise, individuals

likewise discover vehicles more reasonable. In any case, this movement

of vehicle reception regularly raises emanations and gridlock in the

climate [14]. Our reality facilitated 1 billion vehicles in 2010, and this

ﬁgure is extended to twofold by 2030, creating a quick requirement for

extra ﬁnancing and backing for foundation to deal with this huge ascent

in the quantity of vehicles [15]. We have numerous issues or issues today,

for example, not having a parking spot, carbon contamination, an enor-

mous number of auto collisions (or genuine wounds/fatalities) [16].

Around 1.25 million individuals kick the bucket every year because of

street mishaps, as per the World Health Organization (WHO), and the

WHO has assessed that the loss of life could increment to 1.8 million by

2030 [15]. The requirement for mechanically progressed, completely

mechanized, dependable and stable methods for transport is hence sig-

niﬁcant, and the self-governing keen vehicle industry has been endeav-

ouring to satisfy these guidelines.

Be that as it may, improvement in the notoriety of Autonomous Ve-

hicles has been expanded because of some prominent mishaps before,

while vehicle makers actually do their most extreme to determine the

applicable/fundamental issues. In any case, duties made by the car

business, for example to completely showcase Autonomous Vehicles

before 2035–2040, are hard to make. Autonomous Intelligent Vehicle

(AIV) is additionally a long way from our thought that almost 2050 may

happen. Self-sufﬁcient Intelligent Vehicles are the need of things to come,

the vehicles of things to come (or the vehicles of things to come), which

will improve the way of life of people or the lifestyle. As talked about in

Ref. [17], self-sufﬁcient vehicles will keep on having numerous signiﬁ-

cant issues, for example, government guideline, consumer loyalty, fore-

cast of market immersion, cost, dependability and assurance, which must

be completely tended to by the pertinent gatherings associated with the

self-sufﬁcient savvy vehicle area. In outline, due to its underlying sig-

niﬁcant expenses and helpless dependability, self-sufﬁcient clever

vehicle innovation will require some investment before it gets useable

and reasonable to purchasers. However, vehicle producers must address

these issues (requiring airbags, i.e., wellbeing things) before they can

make Autonomous Vehicles a triumph and change the car business.

Scope of this work: We present an exhaustive and deliberate

investigation of best-in-class results for self-sufﬁcient insightful vehicle

innovation in this paper. We are examining existing independent canny

vehicle innovation arrangements, including their design, executions,

testing and conﬁrmation. Present issues and difﬁculties in the organiza-

tion and making of self-governing shrewd vehicles (for genuine clients)

are likewise tended to top to bottom. We are likewise centred around the

plan and use of self-ruling canny vehicle innovation and tending to this

inside and out. What's more, we analyse both innovative and non-

specialized usage issues for Autonomous Vehicles that must be exam-

ined in the independent shrewd vehicle improvement chain by all part-

ners/enterprises. The critical commitments to this paper are summarized

underneath:

a) We present a nitty gritty and orderly examination of self-ruling canny

vehicle innovation tending to issues of plan and execution.

b) We distinguish cutting edge discoveries from both business and

scholastic viewpoints on independent vehicles.

c) In Autonomous Vehicles, we clarify plan and usage issues in detail.

d) We present a top to bottom review of the various examination issues

(specialized, non-specialized, social and political) that the indepen-

dent canny vehicle industry needs to illuminate.

e) To put it plainly, this paper offers an itemized writing review on late

examinations directed in the ﬁeld of self-sufﬁcient insightful vehicle

innovation and overcomes any issues between the difﬁculties looked

via self-sufﬁcient vehicles in design, usage and exploration.

Also note that Autonomous technology means the car which do the

task with minimal input form the driver or from its own, for example,

Auto-parking is an autonomous technology to park a vehicle. Here

autonomous features can be varying into Semi-autonomous and fully

autonomous. Semi-autonomous means the responsibility for driving falls

to the driver and the technology is used in the car as a driver-air and

safety feature, for example, Tesla cars. On another side, fully autonomous

means, the car is able to fully drive itself like Herbie the Love Bug or the

Batmobile and operators (rather than drivers) are able to give full control

and responsibility to the machine. This type of cars is still is yet to come

in near future with 100 % accuracy.

Further, Fig. 3 explains the organization of our present work in detail.

This present paper's key examination commitment is to analyse the

current cutting edge of self-sufﬁcient vehicles and to investigate a portion

of their issues and likely conceivable outcomes. We likewise perceive

applications that can prompt improving the trafﬁc difﬁculties confront-

ing the present metropolitan culture. In this work or vehicle's scenarios,

the words "self-driving", "driverless," "self-governing," and "computer-

ized" or "self-ruling vehicles" and "independent vehicle (with knowledge)

or autonomous vehicles" are used interchangeably. Recall that the key

distinction is the capacity to play out the function of driving. 'Self-driving'

is an expression used to indicate a vehicle's capacity to screen at least one

of the highlights of the vehicle as it is moving starting with one phase

then onto the next. The essential perspective is that with their hands on

the guiding wheel, an individual must be in the seat. The expression

"driver less" is utilized to allude to vehicles that track their encompassing

climate and react safely dependent on the sensor recovered data, i.e., it is

not fundamental for a human to be steering the ship.

A.K. Tyagi, S.U. Aswathy International Journal of Intelligent Networks 2 (2021) 83–102

Organization of the work: The rest of the paper is organized as

follows: Section 2covers emergence of Autonomous Vehicles with

Existing Reviews for the same. Further, section 3discusses our motiva-

tion behind writing article for this area/transportation sector. Then ob-

jectives, types, classiﬁcation and importance of AIV are discussed in

section 4. Further, major beneﬁts of Autonomous Intelligent Vehicles will

be discussed in section 5. Design and Implementation of autonomous

intelligent vehicles will be explained with a detailed explanation in

section 6. Further, section 7will discuss several open issues and critical

challenges towards faced implementation of AIV for real world. Then,

various research opportunities will be included and discussed in section

8. In last, section 9will conclude this work with including several

interesting remarks for future researchers/readers. Also, for more details

about work, all abbreviations used in our work have been listed in

Table 1.

2. Emergence of autonomous vehicles with Existing Reviews

The principal driverless vehicle endeavour goes back to the mid-

1920s [18] and picked up foothold during the 1980s when specialists

prevailing with regards to making mechanized interstate frameworks

[18]. This prepared for the connection of semiautonomous and

computerized vehicles to the thruway framework. From 1980 to 2000,

pioneer AV pilots were generally created in Germany and the U.S [19].

AVs are incredibly obligated to the thorough exploration directed by the

security area known as the U.S. (DARPA) on automated gear, an Orga-

nization for Advanced Research Projects in D

efense [18]. The driverless

vehicle from Google outﬁtted the AV with enormous advertisements and

pulled in a pool of ability from numerous controls. Google's driverless

armadas logged over 1,000,000 miles as of late as July 2015, during

which just 14 minor public street auto collisions were accounted for.

Notwithstanding, the AV was not to blame in all cases; rather, it is

possible that it was driven physically or the other driver was to blame

[19]. In any case, on Valentine's Day 2016, when the vehicle hit the side

of a public transport in the Silicon Valley town of Mountain View [18],

the principal crash where the Google vehicle was found to blame

happened.

2.1. Emergence of autonomous vehicles

It is anything but a direct street to a driverless future. It needs a scope

of extraordinary innovations, yet in addition a solid (or severe) lawful

structure and a novel protection model to convey self-ruling vehicles out

and about and increase public acknowledgment. To move to a driverless

climate, assistive advancements (for example journey control, park help,

versatile voyage control, and path keeping help) are more signiﬁcant. For

vehicle producers and providers to fabricate fabricating plants outﬁtted

with the vital innovation, to create groups and abilities to popularize

independent vehicles, the appropriation and joining of these components

into new vehicles will be signiﬁcant. Vehicle makers, for example, Volvo,

BMV, Audi, GM, Toyota, Ford, and Mercedes Benz, and so forth, are

actualizing/coordinating help frameworks in their vehicles/vehicles as a

methodology to empower drivers to begin grasping and depending on

innovation that guides them and supports their driving decisions.

Scarcely any vehicle producers have pledged to present driverless vehi-

cles in 2019/2020, albeit few makers, for example, Honda have reported

that they will sell semi-computerized vehicles by 2020 and completely

robotized vehicles by 2025. Nonetheless, it isn't workable for indepen-

dent vehicles to hit this imprint in the following ﬁve years because of the

serious presence and high portability of vehicles. In any case, in the

coming years, the selection timetables for self-governing vehicles will

rely upon administrative laws and the acknowledgment of help frame-

works. Investigation of Autonomous Vehicle Stakeholders: We can

analyse a few businesses or partners for self-governing vehicles or make

AV savvy (additionally mechanized vehicles. Such referenced partners

have a section in AIV's presentation.

Fig. 3. Key organization of our work.

A.K. Tyagi, S.U. Aswathy International Journal of Intelligent Networks 2 (2021) 83–102

Vehicle Manufacturers

Road Users

Insurers

Transportation Technology Managers

Note that Transportation Technology managers are crucial factor in

the driverless transition phase. It is the implementation of infrastructure

to enable the use of autonomous cars. We need to develop, however a

strategy to incorporate autonomous vehicles into the existing in-

frastructures of roadside and communication. In order to better handle

infrastructure maintenance and planning, transport authorities need to

update the digital infrastructure so that 5G, CV2X, and Wi-Fi can help.

Furthermore, in order to be able to exchange data in a secure and open

manner, transportation agencies would have to strengthen their data

storage and communications networks while generating processes, pol-

icies and agreements.

Various audits have been completed to date that talk about various

parts of independent keen vehicle innovation [20–30]. The vast majority

of these studies, nonetheless, focus on just a single part of the indepen-

dent canny vehicle supposedly, and there is no study that gives an efﬁ-

cient way to deal with the self-governing shrewd vehicle, self-sufﬁcient

clever vehicle innovation. We are utilizing Autonomous Intelligent Ve-

hicles (AIV) unexpectedly, to our full information. There is no such word

on the web, just terms for clever vehicles or self-ruling vehicles bring

about list items. There are papers from the most recent 10 years

(2011-date) in our examination. In this, Campbell et al. [20] examined

this present reality self-governing keen vehicle tests in metropolitan

conditions and recognized in detail the difﬁculties looked during test

drives. A thorough overview on the variation of Advanced Driving

Assistance (ADAS) in Autonomous Vehicles was performed by Okuda

et al. [21]. Fagnant et al. [22] reviewed the Autonomous Vehicles

strategy rules and execution. Also, Bagloee et al. [23] tended to a portion

of the troubles related with different self-governing insightful vehicle

arrangements. Other capacity explicit reviews incorporate the arranging

and movement control of self-ruling vehicles [165, the improvement of

long-haul maps for self-ruling vehicles [27] and visual impression of

independent canny vehicles from the perspective of both usage and cli-

ents [29,30]. Also, a study on client certainty and their desires for

self-sufﬁcient astute vehicle innovation was led by Abraham et al. [25],

while Joy et al. [26] investigated issues of availability and security in

self-sufﬁcient vehicles. Parkinson et al. [28] have investigated digital

weaknesses in Autonomous Vehicles thoroughly. We note that latest

Autonomous Vehicles reviews have focused more on speciﬁc subjects of

the independent shrewd vehicle from the past conversation. We ﬁnd that

no creator has endeavoured to make independent canny in this exami-

nation, yet zeros in additional on delivering the vehicle driverless.

3. Motivation

Autonomous Intelligent Vehicles or Intelligent Transport Systems

(ITS) are getting consideration from past quite a long while. As of late

because of street crashes, a great many people have lost their lives or

have been for all time incapacitated around the world. This has given the

occasion to go to an independent vehicle since human mistakes are

answerable for practically 50 % of the car crashes. As per gauges, the

quantity of lives guaranteed by trafﬁc episodes every year is relied upon

to twofold in the following 10 years [31]. Every year, about 1.24 million

individuals bite the dust because of street auto collisions, as indicated by

the WHO. The main source of death among youngsters matured 15–29

years, is street car crashes. "Weak street clients" are half of those executed

on the world's streets: walkers, cyclists and motorcyclists [12]. Street car

crashes are relied upon to bring about the passing's of about 1.9 million

individuals every year by 2020 without activity [32]. Note that

completely Autonomous Systems/Autonomous Intelligent Systems is

relying upon full mechanization and progressed examination for settling

on choice absolute toward the end in parts of second for keeping away

from any fatalities or mishap over the street/roadway (during driving).

Driverless is the future for mankind in 21st century, need to prepared by

human and smart and computerized frameworks.

4. Autonomous intelligent vehicles: objectives, types,

classiﬁcation and importance

Today, a few vehicle makers are taking a shot at tasks to manufacture

savvy vehicles, which are displayed each year at engine shows (CES,

Geneva Motor Show, London Motor Show, and others in signiﬁcant

urban communities. Self-ruling vehicles are viewed as a normal answer

for lessen the quantity of auto collisions, clog expenses and levels of

discharges, while expanding the effective utilization of time spent in a

vehicle. We are encountering an advancement in the vehicle business

(from around the globe) in this century. The improvement of more

intelligent vehicles (counting self-governing vehicles) is being quickened

by ongoing specialized turns of events and the fast expansion of in-

novations. We address self-governing vehicles in the above areas,

depicting the essential partners and their function in their prosperity or

disappointment.

4.1. Objectives of autonomous vehicles

It is conceivable to depict a completely self-ruling vehicle as a vehicle

that can see its current circumstance, ﬁgure out which course to take to

its objective, and drive it. Brilliant vehicles or robocars are self-governing

vehicles that utilization a mix of sensors, PC processors, and information

bases, for example, maps, to assume control over a few or the entirety of

Table 1

Abbreviations used in our Work.

Short Forms List of Abbreviations

AIV Autonomous Intelligent Vehicles

ITS Intelligent Transport System

IoV Internet of Vehicles

CCTV Closed-Circuit Television

ALV Autonomous Land Vehicle

VANET Vehicular Ad Hoc Networks

OBU On-Board Device

DSRC Dedicated Short-Range Communication

IoT Internet of Things

V2V Vehicle to Vehicle

ADAS Advanced Driving Assistance

NHTSA National Highway Trafﬁc Safety Administration

LIDAR Light Detection and Ranging

VLC Visible Light Communication

GPS Global Positioning System

ECU Electronic Control Units

CAN Controller Area Network

MaaS Monitoring as a Service

CPU Central Processing Unit

CNN Convolutional Neural Network

DNN Deep Neural Networks

DRL Deep Reinforcement Learning

FCN Fully Connected Network

DBM Deep Boltzmann Machines

ML Machine Learning

DL Deep Learning

CRF Conditional Random Field

HCRF Hybrid Conditional Random Field

V2I Vehicle to Infrastructure

DGC DARPA Grand Challenge

VIAC VisLab Intercontinental Autonomous Challenge

VoE Vehicle of Everything

V2R Vehicle-To-Road-Side Unit

V2P Vehicle-To-Pedestrians

V2D Vehicle-To-Devices

V2N Vehicle-To-Networks

V2G Vehicle-To-Grid

V2H Vehicle-to-Home

PLC Programmable logic controller

RADAR Radio Detection And Ranging

A.K. Tyagi, S.U. Aswathy International Journal of Intelligent Networks 2 (2021) 83–102

the human administrators' driving capacities. Vehicles that are ﬁtted with

this innovation would have their own points of interest. It is probably

going to limit wounds, energy utilization and drastically lessen dis-

charges. Major OEMs have as of late uncovered their aims to begin selling

such vehicles inside a couple of years from now. The independent vehi-

cle's key destinations are portrayed beneath:

a) Perception

b) Motion Preparation - (direction) steering, pace

c) Navigating

d) Behaviour-study of lanes, overtaking

The points mentioned above are targets for real-world users of

autonomous vehicles.

4.2. Classiﬁcation of autonomous vehicles

The National Highway Trafﬁc Safety Administration (NHTSA) has

typically categorised autonomous vehicle technology into 6 stages [32]:

a) Level 0 - no computerization; all undertakings are performed by the

driver.

b) Level 1 - Automation/driver help fundamental capacity: This would

incorporate the robotization of explicit control capacities, for

example, journey control, path direction and equal programmed

stopping. Drivers are completely drawn in and responsible (hands on

the guiding haggle on the pedal consistently) for by and large vehicle

power. Note that the vehicle is driver-controlled, however there

might be driver help attributes, most vehicles out and about today

have some driver help qualities, for example, voyage control,

vulnerable side location or park help.

c) Level 2 - Hybrid Function Automation/Partial Automation: This im-

plies the computerization of various and coordinated control capac-

ities, for example, path focus versatile voyage control. Drivers are

responsible for street oversight and are needed to be accessible for

control consistently, yet might be separated from vehicle adminis-

tration under certain conditions.

d) Level 3 - Restricted Self-Driving Automation/Conditional Automa-

tion: Drivers can, under certain conditions, surrender all wellbeing

basic capacities and depend on the vehicle to follow changes in those

conditions that will require a re-visitation of the driver's control. It

isn't normal that drivers will effectively follow the street.

e) Level 4 - Complete Self-Driving Automation/High Automation: Ve-

hicles are ﬁt for playing out every single driving capacity and

following street conditions for the whole excursion, as are equipped

for working with non-driving tenants and without human inhabitants.

f) Level 5 - Fully self-sufﬁcient: the vehicle is ﬁt for playing out all ca-

pacities under all circumstances, and the driver may have the alter-

native of controlling the vehicle, yet at this level no driver is required,

and no guiding wheel might be required.

Now, vehicles can be differentiated into several types:

Internet of Vehicles

Autonomous Vs Automotive

Connected Vehicles

Semi-autonomous vs Fully autonomous

Autonomous Intelligent Vehicles

Hybrid Electric Vehicle

The Autonomous Vehicle/Vehicle

As discussed above, autonomous vehicles mean autonomous vehicles,

light weight vehicles like ambulances, etc. Automated Ambulance Ve-

hicles (a part of Internet of Vehicles) in near future can change the e-

heath vehicle. Many live can be saved around the globe annually. In this

work, we explain autonomous intelligent vehicle as popular example of

autonomous vehicles. Sometime, we may refer use of Automated Vehi-

cles in e-healthcare vehicle applications.

Commercial vehicles or autonomous vehicles will bring a lot of

change in the near future, i.e., a lot of real-time data (vehicles equipped

with many sensors and actuators) will be produced that must be pro-

cessed and evaluated in order to receive timely decisions to beneﬁt

people/users/citizens. Autonomous vehicle design should be such that

during driving, moving over the road or street, no error is created. If

something happens or an obstacle comes immediately in front of any car,

it can make timely decisions. But today, for autonomous vehicles, these

are the major challenges, i.e., immediate and timely decisions are too far

from reality. The autonomous intelligent vehicle comes into the picture

for certain problems (in which all functions/operations operate inde-

pendently). In general, the volume, speed, consistency, heterogeneity

and real-time nature of the data must be taken into account in the design

of an autonomous vehicle. Note that on-board sensor and actuator ad-

vances are utilized by different auto (or vehicle) producers for different

kinds of streamlined applications. In outline, the primary necessity of

future innovation is independent astute vehicle plan. As such, the self-

sufﬁcient savvy vehicle requires highlights that will permit it to

securely foresee, decide, and move and give end clients/residents with

solid assistance.

The key signiﬁcant level useful parts of a common independent

vehicle framework are appeared here in Fig. 4, where every segment has

its focal points (experts) and constraints (cons). The layered structure

incorporates equipment segment information procurement, for example,

on-board and in-vehicle sensors; short and long-range radars; Light

Detection and Ranging (LIDAR) following; and handsets (cameras and

specialized gadgets). The information acquired by these segments is

handled by the focal PC arrangement of the independent vehicle, and is

then utilized by the choice emotionally supportive network (additionally

utilized by numerous organizations to enhance existing models) to

improve the current framework and more secure. Situational information

is acknowledged in self-governing keen vehicles by both short and long-

range imaging sensors that incorporate radar, LIDAR, and cameras. For

different applications, various scopes of situational mindfulness apply,

and they are reﬁned by different segments. The independent vehicle is

regularly ﬁtted with a few cameras to give perspectives on the general

climate, while LIDAR is utilized to forestall impacts and crisis slowing

down. Then again, long-range radars accomplish co-useable voyage

control and long-range trafﬁc see development. For instance, through a

progression of steps, a self-sufﬁcient vehicle moves from point A to point

B: the vehicle needs to see and get mindful of the outer climate, plan the

excursion, explore, and make controlled movements out and about.

Subsequently, a portion of the above assignments are now referenced

that impressively affect Autonomous Vehicle development or enactment,

included here as:

a) Knowledge of situation and environment;

b) Planning navigation and paths; and

c) Power of manoeuvres.

d) Other

We note that several iterations are performed for a vehicle, i.e.

starting from the source until it reaches the destination. The vehicle

needs to gather surrounding scenarios and environmental knowledge of

the world after looking around the environment to make vehicle move-

ment effective/useful for people. Note that the location of the vehicle can

be tracked by navigation. In addition, autonomous smart vehicles

communicate with many other environmental vehicle agencies,

including roadside networks, neighbours (Autonomous Vehicles and

Connected Vehicles), registration and management bodies and service

providers. How these elements and entities interacted with each other is

addressed in Ref. [2]. In design, connectivity, software, and facilities,

today's connected vehicle technology has produced impressive results.

Associated vehicle (non-self-ruling vehicles) innovation is acknowledged

A.K. Tyagi, S.U. Aswathy International Journal of Intelligent Networks 2 (2021) 83–102

by Vehicular Ad hoc Networks (VANETs) where street vehicles speak

with one another through different fundamental correspondence ad-

vancements (for example IEEE 802.11p, WiFi, LTE, Visible Light

Communication (VLC, etc., with the foundation and with the climate.

Recall that the very correspondence conventions that are utilized in

associated vehicle innovation will likewise be utilized for self-sufﬁcient

shrewd vehicles. This would not just energize the assembly of the

self-ruling keen vehicle with the ﬂow half and half electric vehicle

(insight containing) or associated vehicle innovation, however will

likewise encourage the sending of self-governing shrewd vehicle corre-

spondence. Notice that man-made consciousness moves human knowl-

edge here and renders vehicles self-governing, insightful, steady, and

secure.

4.2.1. Situational and environmental awareness

By and large network information, for example object following, self-

situating, and path spotting, is required via Autonomous Vehicles. All in

all, by utilizing equipment modules, going from ready and on-vehicle

cameras to medium and short-range radars, vehicles need to see what

is before them, for example 360-degree network observation (without

loss of over-simpliﬁcation). Then again, cameras are valuable for infor-

mation on the climate and the network (for more data, allude to Fig. 2).

Notice that for vehicular registering/vehicular distributed computing,

the volume and speed of constant information required for neighbour-

hood mindfulness is excessively troublesome and complex. Likewise, the

granularity of information gathered from cameras is conversely relative

to the speed and productivity of the choice emotionally supportive

Fig. 4. High-Level Functional Parts of a standard Intelligent Autonomous Vehicle System.

Fig. 5. Current and future challenges for Autonomous Intelligent Vehicle.

A.K. Tyagi, S.U. Aswathy International Journal of Intelligent Networks 2 (2021) 83–102

network. Radar innovation is utilized as a compelling item following

innovation for AIV contrasted with cameras, settling on it a more prac-

tical decision for self-governing vehicles. Note that LIDAR following is

utilized via self-governing savvy vehicles, i.e., LIDAR gives 360-degree

representation and item following (for long reach). Notice that the

LIDAR framework can likewise be put on top of the vehicle to get an away

from of the zone, yet it (LIDAR) doesn't turn out proﬁciently for serious

item discovery, for example, crash opposition during leaving, impact

evasion, and guard wellbeing. Advanced radars are mounted at the front,

back and sides of the vehicle for such conditions. Gathering information

from the sensor utilized by specialists to enable the choice to emotionally

supportive network of the self-governing vehicle looks after speed, apply

brakes, switch to another lane, make a 3D picture of the general climate

and explore (to make future/future choices).

4.2.2. Navigation and path planning

Route or direction on the ideal bearing in a self-sufﬁcient savvy

vehicle utilized for making/voyaging. On the off chance that the self-

sufﬁcient smart vehicle knows about its environmental factors, its

course should be modiﬁed (with less blunder) in light of the objective.

Note that independent astute vehicles attempt to locate the most secure

yet not ideal course to arrive at the objective. The self-sufﬁcient vehicle

ascertains a course between the current area and the objective with the

guide of route equipment, for example, the Global Positioning System

(GPS) module. The route arrangement of the vehicle (GPS: exactness,

advanced and reduced equipment, on-chip conﬁguration, ease) is utilized

today by billions of vehicles. The self-governing savvy vehicle direction

framework occasionally tests the development of the vehicle against the

deliberate way during the way estimation, while street networks are

actually pre-characterized. Notice that GPS-based arrangements offer

improved direction and route usefulness, however don't offer solid types

of assistance for common or fake wonders, for example, submerged

streets and passages, so wise vehicles utilize inertial direction and route,

(for example, whirligigs and accelerometers). On account of Autonomous

Vehicles, when the importance of development is perceived, both the

whirligig and GPS will function admirably together. GPS data is ordi-

narily utilized for Autonomous Vehicles as a contribution to a unique

calculation for map age that utilizes information securing and tactile data

got from the vehicle.

4.2.3. Manoeuvre control

It starts its excursion after the self-ruling shrewd vehicle sees its

environmental factors and uses this information alongside its objective

data. Note that self-ruling keen vehicles pick the way of the monster from

source to objective, not the ideal course. A vehicle utilizes different

moves/sensors/actuators/direction frameworks for a completely

controlled, smooth, secure excursion along the path. Note that vehicle

segments are controlled electronically through electronic control units

(ECUs), for example ECUs speak with one another and by means of the

Controller Area Network (CAN) transport inside every vehicle with the

choice emotionally supportive network. Along these lines, during the nay

ride, the independent canny vehicle (or self-ruling knowledge vehicles)

must keep up path keeping, heavily congested separation, abrupt brakes,

overwhelming, and halting at trafﬁc signals. These moves need help for

equipment/programming and broad correspondence and constant

sharing of information between the different control frameworks of the

vehicle. Note that different types of moves include path keeping; packed

in separation; unexpected brakes; surpassing; and halting at trafﬁc

signals.

4.2.4. Others

Way arranging, object recognition, and so forth, are signiﬁcant seg-

ments that make an AIV all the more remarkable and productive, as

talked about above. Hence, specialists are keen on the ﬁeld of indepen-

dent computerization, incorporating robotization with knowledge, and

much has been cultivated here, as talked about in this part. We found that

complete organization driven frameworks related to vision guided

highlights is the fate of independent vehicles. Toward the start of the

following decade, most organizations are intending to convey completely

self-governing vehicles. An aspiring time of sheltered and helpful

portability is the eventual fate of self-sufﬁcient vehicles. New corre-

spondence and advanced mechanics developments have huge affected

our regular way of life, with transportation being no special case. These

advancements have offered ascend to the possibility of Autonomous

Vehicle (AV) innovation pointed toward lessening impacts, energy uti-

lization, outﬂows and clog, while expanding transport productivity.

5. Design and Implementation of autonomous intelligent

vehicles

Assurance, strength, elegant crumbling, safeguard nature, equip-

ment/programming plans, and consumer loyalty will decide the eventual

fate of Autonomous Vehicles. The plan and sending of self-ruling vehi-

cles, nonetheless, must give extraordinary accuracy, assurance and

dependability to accomplish these objectives, since human lives depend

straightforwardly on them. The AIV is centred around signiﬁcant in-

novations, for example, LIDAR, radar, situating, sonar, progressed sen-

sors, and programming advancement. Table 2 examines a portion of the

dialects utilized based on the test system and different subtleties.

6. Major beneﬁts of autonomous intelligent vehicles

The idea of the self-sufﬁcient insightful vehicle, notwithstanding its

multifaceted nature, opens up new inventive applications and presents

purchasers with wellbeing, convenience, solace, and worth added ad-

ministrations. This part examines advantages of self-governing vehicles/

self-sufﬁcient savvy vehicle in not-so-distant future.

6.1. Improved safety

In car applications, security is the most elevated need segment ne-

cessity particularly in transportation frameworks (since human life is

viewed as the most elevated need). A large number of lives are lost yearly

in the transportation area (everywhere on the globe) [32]. Consistently,

numerous street mishaps lose 1.3 million lives and 50 million genuine

wounds over the globe, as talked about in the motivation portion. Recall

that, because of human mix-ups, many auto collisions happen. Different

factors including interruption, forcefulness, vehicle lessness, inebriation,

and incapacities, cause by human errors. The large transportation issue is

sheltered driving. What's more, government spends gigantic wholes on

their kin because of these slip-ups, for example in improving wellbeing

expenses and punishments (likewise executing hardware over street to

make travel/development more secure). A self-governing astute vehicle

will likewise be a superior choice later on (as elective human-driven

vehicles) that will have less mishaps. The vehicle itself is another

component of assurance. To verify its genuine clients, the self-ruling

vehicle (with insight) will be ﬁtted with trend setting innovations, in

this way keeping crooks from taking the vehicle. The self-sufﬁcient wise

vehicle can recognize its legitimate proprietor effectively with innovative

sensors ready and it sends the proprietor an admonition in the event of

any unforeseen condition. These highlights are just incompletely acces-

sible in existing mid-and top of the line vehicles. Notice that the level of

insight in future self-ruling vehicles will increment altogether. Moreover,

as traditional vehicles, a self-governing canny vehicle needn't bother with

a key to begin. With biometrics, for example, ﬁngerprints, a retina

examine, voice acknowledgment programming, as well as manufactured

clairvoyance, Autonomous Vehicles may work. There is a unique mark

empowered entryway lock framework in current vehicles today, however

these vehicle activities have not yet developed enough to a level that

biometrics can be utilized for a bigger scope or for business use).

A.K. Tyagi, S.U. Aswathy International Journal of Intelligent Networks 2 (2021) 83–102

6.2. Privacy protection towards autonomous intelligent vehicles

In this new (smart) age, location privacy and identity privacy are two

major issues in Autonomous Vehicles and Connected Vehicles. Several

authors [33] in have attempted several attempts to maintain these types

of privacy over the road network, but it is very difﬁcult to preserve the

privacy of users/drivers completely due to high mobility features

(decentralised structure) of vehicles. Some research works towards

vehicle adhoc network and future vehicles have been presented by au-

thors in Refs. [34–39]. We request readers/researchers to refer these

works for ﬁnding efﬁcient solutions for preserving privacy in AIV.

6.3. Business opportunities and increasing revenue

Portability as a Service (PaaS)/Monitoring as a Service (MaaS) and

vehicle sharing are two of the energizing innovations made conceivable

without repetitive human communications via Autonomous Vehicles.

Numerous client assets, including capital, time, space, etc., (for example

drivers/travellers), would be spared by the MaaS worldview [40]. Rather

than claiming a vehicle, independent vehicles can be utilized as an asset,

requiring not just a lot of cash ahead of time, yet additionally a driver and

a parking spot. Vehicle sharing (i.e., ride-sharing or carpooling) is pres-

ently a famous client application. With the appearance of Autonomous

Vehicles, be that as it may, carpooling can turn out to be more fruitful by

permitting more proﬁcient utilization of self-governing wise vehicle in-

struments. Carpooling administrations have increased more consider-

ation from ordinary suburbanites throughout the most recent few years

for various reasons, for example, setting aside cash and time (likewise

abstaining from driving difﬁculty, i.e., making clients calm).

With traditional carpooling, when getting individual voyagers on the

course, there are consistently time limitations. Moreover, the cost shared

by the travellers may likewise consider the expense of the driver. We will

wipe out such uses by utilizing Autonomous Vehicles in the present

period for carpooling administrations. This move won't just create

monetary advantages, yet will likewise lessen air contamination in

worldwide metropolitan urban communities exacerbated by trafﬁc cir-

cumstances. It likewise makes immense market openings and changes the

two clients and specialist co-ops' mentality. Self-ruling vehicles would

likewise reform the matter of cabs and vehicle rentals. Taxi specialist

organizations will at this point don't need drivers, decreasing expenses

and raising income. Moreover, with a decreased labour force, rental

vehicle organizations will have the option to smooth out their business

activities. Likewise, due to brilliant applications, (for example, vehicle

sharing, taxi administrations, and lease a-vehicle beneﬁts that are open

by means of individual gadgets, such a change in outlook would likewise

help the tech business. To put it plainly, in the following decade,

mechanized vehicles will help raise deals and reduction work costs.

6.4. Ease of use and convenience

Ease of use and convenience is one of the other beneﬁts that Auton-

omous vehicles provide. There are times when, due to medical disabil-

ities or intoxication, people are unable to drive. For such situations in

addition to the young drivers without a driver's licence, safe and suitable

mode of transportation for the elderly and for people unable to afford

their own vehicles, the concept of autonomous intelligent vehicle can be

put into action. Autonomous intelligent vehicle thus ensures the safety,

cost effectiveness as well as increased mobility rate of citizens.

6.5. Improving trafﬁc conditions

Another aspect of development which autonomous intelligent vehicle

deals with is improving trafﬁc conditions. By increasing the per-vehicle

occupancy, the autonomous vehicles can decrease the count of vehicles

on the road, which can preferably reduce the trafﬁc congestion and

improve the trafﬁc conditions. Human drivers strictly need to maintain

an important parameter, inter-vehicle distance so as to maintain safety

during driving. Autonomous vehicles have the potential to decrease the

distance, making the roads more spacious. Autonomous Vehicles tend to

reduce the trafﬁc congestion on the roads by performing intelligent ﬂeet

management through proper communication with their counterparts on

the roads. Autonomous vehicles have the capacity to control the proper

functioning of trafﬁc laws and thus causing a reduction in the count of

trafﬁc police ofﬁcers on the roads.

Through the selection of best routes [41], Autonomous vehicles also

have the ability to improve fuel efﬁciency, which further reduces the bar

of air pollution. Fuel efﬁciency and the way people drive are directly

proportional parameters. Every driver is different behind the wheel. A

few common driving behaviours are as follows: over-speeding, starting

and stopping, sudden brakes decreasing the fuel efﬁciency and irregular

driving. Autonomous vehicle can be programmed in such a way that they

could be used in a fuel-efﬁcient mode, i.e., rising fuel efﬁciency through

avoiding erratic driving behaviours. Tailgating and unwanted braking

situations which occur on the roads on a regular basis could be avoided

by a proper coordination and communication among autonomous intel-

ligent connected vehicles.

Table 2

Comparison of different trafﬁc simulators.

Simulator Language Scope Mobility Model Features Distribution Platform

Simulink MATLAB In-Vehicle environment In vehicle CAN and XCP communication fusion, driver

models

Commercial Windows,

Unix, Mac

MATSim Java-based

framework

QSim and JDEQSim Macroscopic,

microscopic

Large scenarios public and private trafﬁc fast

dynamic agent-based trafﬁc simulation modular

approach

open source Windows

TRANSIMS Python based run

time environment

Transportation and

individual travel behaviour

Microscopic Simulated very large networks for a long time. Open

source

Windows,

Linux

SUMO C/Cþþ Vehicle –vehicle, vehicle

–anything communication

Microscopic Simulation of multimodal trafﬁc Open

source

Windows,

Linux

Vissim Vissim Com Matlab Planning of urban an Detra-

urban infrastructure

Microscopic Visualization in 2 D and 3D

Virtual design of autonomous vehicles

Commercial Windows

SimLab N/A Vehicle mechanical dynamics NA Simulates different user deﬁned simulation model

Virtual test drive

Commercial NA

veDyna MATLAB Vehicle mechanical dynamics NA Component development and testing

3 D road model, 3 D animation

Commercial Windows

AimSun python scripting Large transport network

(cities)

Microscopic Multiplatform, fast execution, extensible and more

extensible

Commercial Windows

VehicleSim MATLAB, Simulink,

Cþþ

Vehicle mechanical dynamics Mechanical, Vehicle dynamic analysis Commercial Windows,

Linux

A.K. Tyagi, S.U. Aswathy International Journal of Intelligent Networks 2 (2021) 83–102

6.6. Autonomous parking

With the rush of vehicles on the roads, there is one another major

problem which every other metropolitan city faces i.e., the challenge of

parking. This challenge is faced due to the following reasons:

Enormous count of vehicles.

Dense population.

Mismanagement of parking areas.

Inter-vehicular distances in parking lots.

Notwithstanding the development of computerized vehicles, inde-

pendent leaving can somewhat lessen leaving issues. For instance, in the

wake of dropping travellers, self-governing vehicles may leave them-

selves in a thin region, impossible people who might require a bigger

space for a similar activity. This self-ruling stopping thought could spare

an astounding 6.8 billion yards in US parking garages alone.

6.7. Byer-driven methodology

A buyer driven methodology is presented via self-ruling vehicles and

offers drivers the extravagances they need to unwind, kick back and

appreciate the excursion. Clients may appreciate the advantages of uti-

lizing the theatre setup of the vehicle or utilizing the ride time to gauge

their work while driving to their work environment. The formation of

self-ruling vehicles that couldn't in any case be applied is such a traveller

driven methodology [42]. Other such developments can be connection of

a passenger's mobile phone to the autonomous vehicle which might help

the passenger to pick up kids from school, relatives from airport etc.

“TESLA”one of the electric vehicle giant which led to the introduction of

“summon”applications in its high-tech models has witnessed huge

progress in some areas [43]. Summon is a kind of application which al-

lows the user to move to designated places just through a

mobile-applications, to be more pr

ecised they can themselves get parked

in the parking slots and when required the vehicle can move to desig-

nated locations as per the request of the owner. There is add on feature of

this approach which makes it easy for the vehicles to exit the parking

spaces which are very rigid and tight.

More research works are being done to analyse the different driving

patterns and different driver's characteristics and features that mainly

include one's age, gender, driving experience, personality, method of

driving, emotions, history of accidents etc. [44]. The aforementioned

features are what together constitute an individual's driving behaviours.

Knowledge of human behavior forms the base of customization of the

autonomous intelligent vehicles. For example, there are certain factors on

which speeding and overtaking during driving on the roads depends,

which certainly don't limit to characteristics like gender, age, emotion

and preferences. Like, it is generally observed that young drivers are

faster than elders, whereas females and elderly people drive more vehi-

cles fully. One's with the infants and families are generally categorised

under cautious drivers. There is a fraction of population which tends to

opt for less busy roads even if the time taken to reach the destination

might be long. So conclusively, all these statements state that the in-

volvements of all such attributes are necessary for autonomous vehicle's

customization.

One of the most considerate issues is regarding the vehicles when left

alone which essentially implies that they should be speciﬁc about the

places they are supposed to park in and the owner's speciﬁed pick-up and

drop timing. The proprietor of the self-governing vehicle ought to

consider the opportunities for legitimate customization recorded previ-

ously. Self-governing vehicles can likewise plan an approach to explore a

wide scope of traveller jogged applications where vehicles can be

customized dependent on inclinations, for example, speed, hazard levels,

in-vehicle amusement, and so forth These highlights and qualities assume

a critical function in choosing a deﬁnitive driving encounter.

6.8. Others services

In numerous different applications, for example, e-medical care

(rapidly arriving at the objective), coordination's (for proﬁcient circula-

tion of things/bundles), Intelligent Transportation System (ITS) is uti-

lized today. Dazzle individuals could claim a vehicle with complete

voice-actuated ability, and visual and contact gadgets could be utilized

by the individuals who don't talk. The vehicle will perceive the in-

dividuals who need to utilize incapacitated parking spots. What's more,

the vehicle will lead a wellbeing evaluation with the guide of a body

region network if a traveller reports a side effect or indications of an

illness and reports it to the closest medical clinic when driving towards it.

In outline, we can discover numerous helpful advantages through

execution of self-sufﬁcient savvy vehicles for now age.

a) Reduced driver stress: reducing driving stress and allowing motorists

to relax and work while driving.

b) Reduction of driver costs: reduction of costs for taxis and commercial

transport driver's pf paid pf

c) Non-driver independence: offering independent mobility for non-

drivers and thereby reducing the need for drivers to drive non-

drivers and to subsidise public transit.

d) Improved safety: can minimise many common risks of injuries and

therefore crash costs and insurance premiums. Can decrease elevated

rush driving, such as when impaired.

e) Increased road room, reduced costs: platooning (close-up vehicle

groups), narrower lanes, and reduced intersection stops could be

necessary, reducing congestion and roadway costs.

f) More effective parking, cost recovery: It can drop off passengers and

ﬁnd a parking spot, improve comfort for motorists and reduce the

overall cost of parking

g) Increase fuel efﬁciency and decrease pollution: fuel efﬁciency can be

improved and pollution emissions decreased.

h) Shared Vehicles Support: It could promote the sharing of vehicles

(vehicle rental services that replace ownership of personal vehicles),

which can offer various savings.

On another side, advantages and disadvantages of AIVs can be dis-

cussed as:

Safety and crashes

congestion

Taxi and car ownership

Roads' capacity

AV and electric vehicles

Congestion pricing

Value of time

Demand forecasting

Land use

Environment (energy and emission)

Non-industrial nations

Note that among all advantages mentioned above, non-industrial

nations/Third world nations are managing a lack of foundation for

transport, for example, expressways, scaffolds and public vehicle, which

is hampering their ﬁnancial development. These agricultural nations'

appropriation of AVs will spare them the expenses related with the

development of capital-serious foundation. At the point when created

nations jump frogged over to phone advances that absolved them from

expensive landline foundation, a comparative worldview was appeared.

7. Autonomous intelligent vehicles: open issues, critical

challenges

As discussed in above sections and [45], many stakeholders have

invested billions of dollars for making autonomous vehicles reality for

A.K. Tyagi, S.U. Aswathy International Journal of Intelligent Networks 2 (2021) 83–102

people. Tesla is far among all of the stakeholders. In next decade, we can

ﬁnd autonomous cars over the road network (at least with some intelli-

gence but not completely). By 2040–2050, a vehicle with full automation

and intelligence can be required. We faced many problems and crucial

challenges in AIV when implementing or making this reality. This

chapter explores many transparent and important challenges faced by

autonomous intelligent vehicles in this section. A thorough analysis of

Open Issues and Mitigated/Identiﬁed/Noticed Critical Problems in AIV is

presented in Table 3.

7.1. Challenges in computer vison for autonomous intelligent vehicles

(AIV)

Note that PC vision is a major ﬁeld covering themes, for example,

division picture securing, division picture arrangement, and so forth with

some reference to Autonomous Vehicles, the creators focus exclusively

on object identiﬁcation, adjustment, and movement assessment. Protest

identiﬁcation, object following, movement planning is as talked about

above, principal models for self-sufﬁcient wise vehicles. To keep up

different moves, the self-governing astute vehicle must identify both

static and dynamic items. In self-ruling wise vehicles, be that as it may,

object recognizable proof is hard for some reasons, for example, shadows,

comparable articles, lighting conditions, and so forth the hidden calcu-

lations (proposed calculations) ought to accordingly, consider these

factors. Article recognition depends on different sensors, going from

economical cameras to speciﬁc LIDAR and radar. Likewise, for different

purposes and kinds of atmosphere, the self-ruling insightful vehicle needs

sensors. To distinguish living creatures, these incorporate noticeable

light (daytime), infrared sensors (evening time or in faint light), and

warm infrared. Notice that object identiﬁcation, semantic division, and

order approaches perform sensibly better regarding precision, yet their

viability is faulty because of the multifaceted nature of the calculation,

computational overhead caused, inactivity, and absence of adequate

highlights, and the intricacy of physically commanding the plan of

highlights for other robotized components. For object identiﬁcation and

arrangement, profound learning components are in this way signiﬁcant.

Notice that making a 3D picture from a 2D (picture change) is likewise a

basic PC vision highlight that should be incorporated for movement ar-

ranging and activation in the AIV.

7.2. Autonomous intelligent vehicle for real world deployment barriers

There are numerous preferences to self-ruling vehicles, for example,

diminished driver strain, portability for non-drivers, expanded security,

expanded eco-friendliness and to give some examples, decreased

contamination. Yet, there will even now be a few deterrents that will ﬁll

in as hindrances to this current innovation's execution. Four of the

greatest difﬁculties will be tended to progress of time:

a) Cost: In the design of those cars, the companies that test autonomous

vehicles there have paid rather hefty sums. For the AV module, which

is far out of the control of a regular guy, Google itself paid about

$80,000? Once this technology is conﬁrmed, it is anticipated that it

will fall to half of the price, which is still a very huge sum. If the prices

of autonomous vehicles decrease in the future i.e., similar to tradi-

tional vehicles, then average citizens will also be able to purchase and

afford these vehicles.

b) Difﬁculties in innovation/Improving innovation: However, numerous

prestigious vehicle producers, for example, Mercedes, BMW, Audi,

Nissan, and so on have just announced that they will be prepared with

a halfway independent vehicle (Level 3), yet it is as yet far to go as

street conditions are not up to check in numerous nations (like

creating). To assemble the trust where we can rely completely upon

these vehicles, this innovation likewise needs broad examination and

testing. At the end of the day, in certain driving conditions, (for

example, thick mist, substantial downpour, or snowfall), independent

vehicles use hardware, for example, cameras and sensors that may

confront similar issues as human drivers. The test is to create robot-

ized vehicle frameworks with the goal that they can react to every

natural condition. It will even now take another 10–15 years, in our

view, and in the assessment of different specialists.

c) Expulsion of old vehicles: The errand of presenting self-governing

vehicles (soon) is scrap all old vehicles that are not outﬁtted with

an independent module since it makes a great deal of unusual waste

and along these lines diminishes the self-sufﬁcient vehicles' exhibi-

tion and security. In the event that the more established vehicles can

be retroﬁtted, there can be an answer, yet again by taking a gander at

the no conventional vehicles ﬂying on the streets, it would be a

gigantic test. This arrangement anyway will come at a greater

expense.

d) Joblessness issue: While independent vehicles have numerous points

of interest, the joblessness issue is the greatest test we anticipate.

There will be no requirement for drivers on the day when self-

sufﬁcient driving would be totally afﬁrmed. In this way, all the in-

dividuals who procure their work as authorized drivers today won't

have the option to acquire it any longer. Taxicabs, shipping and

marine cargo are the main areas to be affected by the presentation of

independent vehicles.

e) Worry about security and protection: Autonomous vehicles will be

connected through the web/web soon, which could be helpless

against a few genuine digital assaults. In the present age, where

everything is directed by hardware or shrewd gadgets, security and

protection (by programmers) are the most serious issues for these

gadgets. Today, electronic information isn't secure and is defenceless

against abuse of data. Any assailant outﬁt may likewise utilize an

independent vehicle to complete their self-destruction missions.

Additionally, as these vehicles are associated by GPS, anybody can get

the area and it tends to be utilized for some sort of awful plan or their

monetary beneﬁt or extortion. A huge volume of client information

may in this manner be gotten via independent vehicles. The data

could be utilized to create modiﬁed administrations and merchan-

dise, yet our exercises and way of life could likewise be followed, our

vehicles could become spies, and there is a moral measurement to

information, possession and use.

f) Norms and Regulations: In request to authorize the cycle of usage of

self-sufﬁcient vehicles over the street/thruway (for business pur-

poses), policymakers need to make practical principles and severe

standards. Exacting and clear (non-complex) laws ought to be set up

to determine protection issues and to address the security and the

executives of information (individual or private information). Any

psychological oppressor association or some other obscure individual

ought not utilize self-ruling vehicles; such concerns ought to be

settled by the public authority.

g) Vehicle selection: People are hesitant to offer a machine full force, for

example individuals are sure or not certain on machines, particularly

on savvy/electronic gadgets. The test is to portray a move from

"computerized" to "full and canny" driving that is all around

controlled. The key errand is to set up a bunch of new help frame-

works and a structure that decides the degree of appointment of

control in different conditions.

7.3. Some other challenges in AIV will be

7.3.1. Technological challenges

In order for a vehicle to be autonomous and intelligent, as discussed

above, vehicles need sensors to collect road information and a central

processing unit to interpret all the data and make appropriate decisions.

LIDARs, radars, cameras, GPS and ultrasound [46] will be included in the

sensors required; these sensors will be used to identify the environment

of the vehicles and to calculate the distance between the vehicle and

nearby objects so that they can be evaluated and responded accordingly

by the Central Processing Unit (CPU) (see Fig. 4). For example, GPS

A.K. Tyagi, S.U. Aswathy International Journal of Intelligent Networks 2 (2021) 83–102

Table 3

Research and deployment challenges in autonomous vehicles.

Challenges Class Key challenges Possible solutions

Technical

Challenges

Validation/Testing ●No complete set of requirements

●Dynamic and non-deterministic operations

●Complexity of operations

●Mission-critical nature

●Functional division of software/hardware components

●Limited operational concepts

●Inductive referencing and machine learning

●methods

●Fault injection

Safety and

Reliability

●Distance driven in test drives does not determine reliability

●Resemblance to human-level conﬁdence in reliability needs a lot of

resources

●Legislation is vague

●Removal of disengagement function is risky

●Unclear validation cycle

●Deﬁning short-term safe missions

●Deﬁning more fail-safe systems

●Developing sophisticated algorithms for small

●missions

●Employing ML, DL, and AI

Software

Quality

●Huge budget requirements

●Too many unforeseen scenarios

●Autonomous intelligent vehicle and its software represent complex

system

●Fail-safe rather than unpredictable outcomes

●Degradation of functionality when needed

●Goal-oriented software quality testing

Computational

Resources

●Autonomous intelligent vehicle is host to heterogeneous sensors

●Massive amounts of data produced

●Increase in cost

●Real-time data processing

●Redundancy increases reliability as well as Costs

●Graphics Processing Units (GPUs)

●Optimized system on chip

●Agreement on standards to make it more open to research

Security and

hacking

threats

●Autonomous intelligent vehicle operates in networked environment

and is prone to network attacks

●CAN bus (in)security

●Malicious code injection, jamming, fuzzing, and hacking threats

●DDoS attacks

●Separate data security from communication security

●Efﬁcient and effective authentication

●AI-based security approaches

●Security by design

Privacy ●Who stores the data?

●Sharing personal and location data has privacy implications

●Convincing consumers to share personal data

●Conﬂict between privacy and quality of service

●Consumer awareness

●General Data Protection Regulation (GDPR) 28

●Acceptable trade-off between anonymity and quality of

information

Accuracy and

efﬁciency

of object

detection

●Trajectory is not constant

●Real-time object detection is hard

●Limitations of RADAR and LIDAR

●Calibration of detection components

●Use of ﬁnite-state machine for incremental path planning

●Intelligent role-based and contextual cooperative mecha-

nisms among components

Sensors

management

●Data from many sensors must be processed in real-time

●Deep learning algorithms are storage and compute intensive

●Data redundancy, outliers, and granularity from sensors data

●Authenticity of sensed data

●Increasing computation and communication resources

●Crowd-sourcing and crowd-sensing

●Sharing sensors' data across nodes

●Trade-off between number of sensors and efﬁciency of data

processing

Decision making

procedures

●Unpredictable environment

●Human behavior is difﬁcult to realize through a machine

●Optimal decisions are challenging to make

●Difﬁcult to detect fault and malfunctions of the system

●Context-aware object detection and perception

●Situation-awareness

●Context- and situation-aware decision and control

Algorithms

Actuation ●Adaptation to unknown environment

●Actuator saturation

●Wrong input can lead to severe consequences

●Fuzzy and Takagi-Sugeno model for actuation

●Input validation for actuators

No-technical

Challenges

Consumer trust ●Consumers may be reluctant to trust autonomous vehicles

●Testing does not answer all the questions/concerns of consumers

●Lack of universal adequate legislation

●Security and consumers' privacy

●Difﬁcult to mimic human driving behaviour

●US-led legislation initiative for autonomous vehicle

●Startups aiming at increasing consumer trust

●Promoting awareness and incentives

●Promotion of success stories

●Guarantee of fail-safeness

●Limit the number of functionalities

Diversity ●Coping with connected and non-connected vehicles

●Human factor in driving is essential

●Uncertainty in human driving

●Unpredictable behaviour of autonomous intelligent vehicle

towards human drivers

●Environmental diversity

●Implementation of stringent trafﬁc laws

●Detection and isolation of malicious driving behaviors

●Social training

Uncertain

cost

●Software cost is too high

●Maintenance and testing are expensive

●Hardware is too expensive at the moment

●Service subscription, enhanced maps updates, and other costs

●Return on investment for service providers

●Limited functionalities

●Leveled utilities and costs

●Use of autonomous intelligent vehicle as a utility

●Cost-effective business model

●Focus on consumer satisfaction

Operational

robustness

●Real-time decision in unpredictable scenarios

●Crowd management

●Hostile environments

●Fail-safeness

●Driving proﬁles

●Object recognition in real-time

●Learning from surroundings

●Intelligent trafﬁc lights management

Liabilities ●Who will be responsible for accident?

●Who will be insured? Vehicle owner, occupant, or manufacturer?

●Manufacturer might put hidden purchase costs to make-up for their

liability losses

●New business model and new regulations/legislations

●Rethinking of insurance business

●Manufacturer-centric solutions

●Efﬁcient forensic solutions

Recent incidents ●Recent real-life autonomous intelligent vehicle accidents decreased

consumer conﬁdence

●Humans do not tend to trust machines with their lives

●Rethink the race for being ﬁrst in commercialization of

Autonomous Vehicles

●Fail-safeness

(continued on next page)

A.K. Tyagi, S.U. Aswathy International Journal of Intelligent Networks 2 (2021) 83–102

systems and park assistance cameras are now available in level 1 vehi-

cles, but there are many issues with the clear recognition of objects for

level 2 and above, such as fog, snow, road work, complex city driving and

other obstacles that can prevent the vehicle from performing the correct

driving action. Software and algorithms are also the greatest obstacles.

Writing a programme to classify any potential scenario is very difﬁcult.

The vehicle may need the ability to learn and essentially be artiﬁcially

intelligent to react to events for which it has not been explicitly pro-

grammed. Vehicles would have to be able to communicate with each

other (Vehicle to Vehicle (V2V) communication) through a network (any

cloud/edge network) or directly (via temporary vehicle-to-vehicle con-

nections) in order to operate effectively. Vehicles can continuously relay

direction and speed information during journeys, such as deaccelerating

and transmitting signals to a car heading into a trafﬁc jam. As a result,

these signals would be observed by all other vehicles behind them and

their speed would be decreased or drivers would be warned to prevent

any crash/accident. The system is claimed to be capable of minimising

collisions in conditions that do not involve human drivers. To be aware of

speed limits, trafﬁc alerts and path optimisation, vehicles would also

need to connect with the infrastructure (V2I communication). As with

existing technology, there are currently some level 1 cars (available on

the market) that can read road signs. Satellite navigation systems are

plugged into trafﬁc monitoring networks in these cars and can change

routes along with a chosen route when trafﬁc is heavy. To make them

more stable and capable of managing larger data ﬂows, these features

would probably require further investment. We require substantial in-

vestment in autonomous vehicles from governments of several countries,

as well as demanding both V2V and V2I connectivity with the installation

of 5G in these vehicles.

As discussed above, these vehicles can present many hazards, such as

hacking, which is a perfect terrorist weapon. There may be some prob-

lems, such as kidnapping or intentionally causing an accident. The

hackers, who will be responsible for solving such problems, can monitor

any car and do any activity. This will be a relatively new issue for the

automobile industry, but there are issues that are being discussed with

companies such as Apple and Google, etc., in the broader world. The auto

industry needs to invest heavily in cyber protection to avoid these issues.

7.3.2. Impacts on economics and business

With autonomous and smart vehicles, individuals can minimise fuel

consumption, increase the productivity of the occupants of the car, and

reduce accidents and related costs. The time individuals will obtain is a

direct advantage of autonomous driving. With an autonomous vehicle

level 5, people would no longer need to drive, so people would be able to

relax, read or learn a new skill. For those unable to drive a car, autono-

mous vehicles will provide mobility, growing independence for the

elderly, the disabled and the young, while creating more social and

economic opportunities. Car producers or third parties are likely to

supply vehicles for a single trip or for a period of time (e.g., taxis), which

may lead to a decline in vehicle ownership. Therefore, less space will be

required in large cities for car parking, which will be especially beneﬁ-

cial, providing more space for playgrounds, parks and sports ﬁelds. There

will be less focus on hardware and more emphasis on software as the

conventional business model of car makers shifts. With autonomous and

intelligent driving, as discussed above, many job losses will result, i.e., for

truck and taxi drivers. Notice that in various industries, such as software

development and engineering, as well as cybersecurity and the data

management ﬁeld, many jobs could be generated. Autonomous vehicles

would also generate a huge amount of data that cyber professionals will

need to handle and secure. Entertainment and marketing are other in-

dustries that could beneﬁt; the time spent commuting could be replaced

by entertainment services.

7.3.3. Law and government challenges

There are a lot of legal questions about the arrival of autonomous

vehicles, particularly who is responsible for an accident? The driver has

no power in level 5 vehicles and several individuals claim that the

manufacturers, designers and software developers can face liability [47].

The environment where level 5 automation communicates with levels 1

to 4 will also be present, another possible ﬂashpoint as responsibility is

spread after an accident. If the autonomous vehicle is self-learning, it will

be difﬁcult to know if the actions of the vehicle are due to what it has

learned independently (through machine/device communication) or

through its original programming. This raises the issue; if the car has

learned to do it, can we blame the manufacturer? Several countries have

taken steps to establish legalisation for the testing of autonomous vehi-

cles on public roads [47] in Germany, China, California, the UK, etc. If

nations do not make strict and different rules, then car makers can

migrate to other nations where such laws exist. Completely autonomous

driving (level 5) would require that national law and international law be

amended.

7.3.4. Ethics and public perception

Public awareness is one of the greatest obstacles for autonomous

vehicles, if autonomous vehicles are branded disruptive as genetic en-

gineering, which took years before proper discussions were feasible, then

autonomous vehicles would not advance, illustrating that public opinion

opposition will bring technology to a standstill. In order to prevent this

from happening, industry and government will have to freely and frankly

discuss the concerns surrounding autonomous vehicles. The cyberse-

curity issues could alter the understanding of people and make some

reluctant to accept the new technology, it is likely that there would be a

Table 3 (continued )

Challenges Class Key challenges Possible solutions

●All unforeseen scenarios are not possible to cover ●Minimise the damage

●Open-source software development

Social Challenges Human

Behaviour

●Humans are generally reluctant to change their behaviors

●Drivers' behaviors towards autonomous intelligent vehicle can be

aggressive

●Autonomous intelligent vehicle may not fully mimic human

driving and thinking behaviour

●Educate people about the technology

●Service providers should provide training/tutorials to the

public

Ethical and moral

consequences

●Right-of-way is more complicated in case of autonomous vehicle

●Human empathy cannot be implemented in autonomous vehicle

●Trolley problem

●Hard to make optimal decisions

●Social equity concerns

●Fast intelligent and real-time response with efﬁcient soft-

ware will mitigate such cases

●Implement multiple ethical theories and test

●Implement different driving behaviors

●Alternate jobs for people who will lose their jobs

Policy

Challenges

Policy

Challenges

●Re-examine many policies may open up further policy challenges

●No clear policy since Autonomous Vehicles are not commercialized

yet

●Safety and utility are inversely proportional

●No clear policy for autonomous intelligent vehicle certiﬁcation

●Policies are already being implemented

●Recommendations that encompass the concerns of all the

stakeholders

●International task force comprised of all stakeholders to

develop sound policies

Note that a relationship about challenged faced in AIV implementation for real world has been depicted in Fig. 5.

A.K. Tyagi, S.U. Aswathy International Journal of Intelligent Networks 2 (2021) 83–102

divide between those willing to embrace the technology and those who

refuse. Ethics will be a crucial subject for the creation of autonomous

vehicles; the trolley experiment, which is based on a theatrical thought

experiment, is an example of this. Is a car meant to swerve into a few

people or a large number of bystanders? Notice that the majority of in-

dividuals will save animals and save the lives of many rather than a few

[32]. If level 5 automation happens and is common, these types of

problems may arise at any time and pre-programmed algorithms will

have to make the decision; ethics will need to be programmed in advance

and individuals may make these moral choices. This could potentially

lead to legal problems as the question arises: will the person who pro-

grammed the ethics be responsible for it? Unless the software knows

ethics itself, is the manufacturer responsible for raising the issue again?

This gives stakeholders/car manufacturers a strong start to step forward

with their plans, but further legislation would have to be given, both

national and international.

8. Research opportunities for future in autonomous intelligent

vehicles

This portion tends to most recent examination discoveries got in the

ﬁeld of savvy self-ruling vehicles and innovation for brilliant vehicles. In

this, we attempt to investigate the specialized side of the self-governing

smart vehicle, segments of programming, ideal calculations utilized over

the street network for PC vision, profound learning, correspondence and

control, and so on Yet, ﬁrst to test the feasibility of Autonomous Intelli-

gent Vehicles, we will examine a couple of certiﬁable tests. Somewhere in

the range of 1990 and 2013, various analyses were completed on driver-

less vehicles for usage in genuine situations [46]. Notice that these tests

were just led on free street (for example without trafﬁc), for example

regarding results/precision we can't demonstrate these tests 'great '. Be-

sides, a few different authors did promote equivalent investigations later

for different situations/conditions. In any case, with close to under-

standing into the activities of drivers of less vehicles, we perceive the

downside of the shrewd vehicles proposed The PROUD test [46] has been

performed and makes signiﬁcant discoveries, for example, the require-

ment for profoundly deﬁnite and solid graphs, a compelling system of

learning and recognition. Moreover, the creators put forth signiﬁcant

attempts with a circulated framework engineering in Refs. [48,49], with

an accentuation on the over-simpliﬁcation of the self-ruling clever

vehicle improvement measure with no reliance on any one-of-a-kind

advancement climate, for example, FlexRay. Note that the Controller

Area Network (CAN) innovation is utilized by customary vehicles for

correspondence between different ECUs. Be that as it may, as a result of

its moderate speed and defencelessness to various assaults, the CAN

transport innovation has been tricky [50]. Afterward, there were not

many endeavours to improve the wellbeing of CAN transports [51,52].

The speed and unpredictability of the CAN transport adjustments keeps

on obstructing its arrangement. FlexRay is to put it plainly, speedier and

all the more impressive, however it's so exorbitant. We discover and at

last, a large part of the independent clever vehicle innovation is exclu-

sive. Hence, we tended to just research discoveries identiﬁed with AIV)

got from specialists having a place with both scholarly community and

industry here.

8.1. Implementation of computer vision in autonomous intelligent vehicles

The essential highlights of Autonomous Vehicles are object identiﬁ-

cation and vision. Self-governing Intelligent Vehicles must see the street

and distinguish any obstruction before and around it, regardless of

whether it is another vehicle, walker, foliage, or some other type of

hindrance, to recognize/identify human driver activities. These two

fundamental highlights make AIV secure, more serious and more reliable.

Note that few helpful errands can be performed by utilizing these high-

lights in an independent (likewise shrewd) vehicle, for example, halting

at a trafﬁc light, easing back down if the past vehicle diminishes speed,

staying away from walkers, etc. Numerous creators have led a few in-

vestigations utilizing PC vision and self-governing wise vehicle object

recognition; however, the discoveries of these tests are as yet not freely

available. Tesla is chipping away at AIV, for example, yet has not un-

covered outcomes because of aptitudes). In Ref. [53], the creators

endeavoured to give a machine overview utilizing PC vision calculations

on astute vehicles. The creators uncover that the recognition, object

identiﬁcation and following, movement arranging, and start to ﬁnish

learning parts of PC vision in Autonomous Vehicles were unequivocally

tended to by the creators. The mistakes made by the most recent PC

vision calculations are ﬂighty and bad in eccentric circumstances, in spite

of signiﬁcant advances in PC vision calculations. The exact exactness of

driving driverless vehicles (independent) securely is exceptionally difﬁ-

cult to recognize. Notice that utilizing social investigation and learning

measures, in unsure circumstances, a choice emotionally supportive

network of the independent wise vehicle will learn and decide, yet we

need a lot of informational indices for this. Computerized reasoning

accordingly assumes a critical function in the independent savvy vehicle

framework's forecast and recognition. Creators reviewed numerous PC

vision calculations utilized for Autonomous Vehicles in Ref. [51], i.e.,

with an attention on path recognition, person on foot and item discovery,

and drivable surface identiﬁcation (utilizing equipment, for example,

GPU, FPGA, and ASIC). The aftereffects of this examination uncovered

that FPGA quickening agents permit FPGA as far as energy effectiveness

and throughput to outﬂank CPU and GPU. Sensor combination, which

joins information from various sensors, is known as a cycle that consol-

idates various sensors to give an itemized, rational and thorough

perspective on the detected information. The work proposed by various

creators in Refs. [54,55] is promising and could be incorporated into

business self-governing savvy vehicle innovation for modern or enor-

mous scope use. Furthermore, in Ref. [56], the creators have proposed a

system dependent on Convolutional Neural Networks (CNN) to identify

3D objects with a solitary monocular camera [56]. To begin with, they

make object recommendations dependent on isolated qualities and af-

terward reﬁne them to characterize genuine articles. When the tactile

information is accessible, it is imperative to arrange the item into clas-

siﬁcations, for example, vegetation, passer-by, vehicle, and so on a cycle

called pixel-level semantic division [57]). Here both AI (managed

learning and solo learning) and profound learning strategies have been

utilized for arrangement purposes to explain this issue. Managed learning

model (for example Support Vector Machine (SVM) for marked infor-

mation), unaided realizing (when information isn't named) can be uti-

lized relying upon accessible tangible information (if named) [58–60]. At

last, profound learning approaches, for example, CNN and auto-encoders

are utilized to improve learning and order measure effectiveness and

mechanize the way toward planning highlights [61,62].

8.2. Implementation of machine and deep learning techniques in

autonomous intelligent vehicles

AI, profound learning and man-made brainpower related methods are

the most remarkable procedures for Autonomous Intelligent Vehicles, as

talked about above and in Ref. [63]. PC frameworks can work naturally

and keenly in numerous enterprises today. AIV can locate the capricious

state and conduct of the encompassing articles (ascertain). Innovation

testing sooner rather than later is likewise upheld in Autonomous Vehi-

cles through AI methods (AI for programming improvement [63]). The

operational rationale is composed physically in ordinary programming

and examined over a progression of experiments, while the product

learns and adjusts with the guide of huge informational indexes in Deep

Neural Network (DNN)- based programming. Next, we talk about a

portion of the normal current profound learning models utilized in

Autonomous Vehicles, (for example, CNN, profound CNN, Completely

Convolutional Network (FCN), DNN, conviction organizations, Deep

Reinforcement Learning (DRL), Deep Boltzmann Machines (DBM), and

profound autoencoders). Notwithstanding vision, other AIV useful

A.K. Tyagi, S.U. Aswathy International Journal of Intelligent Networks 2 (2021) 83–102

standards incorporate scene distinguishing proof, location and

acknowledgment of items (obstructions, vehicles, people on foot, and

vegetation), acknowledgment of human conduct, acknowledgment of the

climate, recognition of street signs, discovery of trafﬁc signals, and

identiﬁcation of vulnerable sides. A DNN-based framework to evaluate

the activities of DNN-driven Autonomous Vehicles was proposed by the

creators in Ref. [64]. While checking the vehicles under different trafﬁc

and ecological conditions, the Deep-Test execution in Ref. [64] discov-

ered incorrect practices a few times. In this work the creators feature the

adolescence of the current arrangements and the requirement for more

severe self-governing vehicle steps to have the option to work completely

autonomously. Note that the current Autonomous Vehicles tests are

sensibly controlled. PC vision [63] is essential for ML and DL strategies,

so PC vision utilizes machine and profound learning methods to perform

different AIV utilitarian segments, for example, object discovery, scene

acknowledgment, location of hindrances, etc.

Moreover, in Ref. [65], the creators proposed a learning instrument to

assess the best possible affordance in AIV that naturally learns various

highlights of an image. Comparable creators have proposed another type

of direct discernment that utilizes CNNs, recognizing key markers of

observation in Ref. [65]. The framework ﬁgures out how to plan different

affordances identiﬁed with driving conduct from a gained picture, for

example, current controlling point, path change, and remaining inside

the path. With TORCS, an open-source vehicle hustling test system, the

creators tried their structure. To recognize street attributes needed for

self-ruling driving, Laddha et al. [66] recommended a cross breed

calculation put together model that works with respect to both directed

and solo learning. A signiﬁcant preferred position of this calculation is

that the creators limited human exertion, for example making AIV

computerized and more versatile, to check the preparation dataset. The

calculation takes different sorts of information from sources, for example,

OpenStreetMap, vehicle-mounted sensors like position and camera sen-

sors. Obstruction discovery, as talked about above, is another huge ca-

pacity of AIV. What's more, profound learning can be successfully utilized

with sensible exactness to identify deterrents out and about. Dairi et al.

[67] proposed a profound learning system dependent on profound

auto-encoders and stereovision to identify deterrents on the path. Also,

different systems have been proposed by outrageous creators and AIV has

acquired noteworthy outcomes. With regards to these amazing outcomes

from a precision viewpoint, it tends to be contended that later on

development of various parts of self-ruling vehicles, profound learning

can assume a basic job. In any case, one of the numerous components that

is hindering the development of these models is the unusualness of the

driving climate.

8.3. Sensors, communications, and control in autonomous intelligent

vehicles

The core of an independent smart vehicle, as examined above is its

registering gadget, which impeccably (possibly) executes the AIV ratio-

nale. For the acknowledgment of a self-governing insightful vehicle

framework, sensors and actuators assume a signiﬁcant job. The self-

sufﬁciency of a self-sufﬁcient wise vehicle implies the taking care of

without human impedance of both known and obscure conditions and

includes learning and counterfeit calculation strategies for example to

check what is best in a particular situation).

These calculations are information concentrated, and the data is

gathered through varieties of various sensors that structure an enormous

sensor network inside the vehicle by and large. Accordingly, information

obtaining, assortment, stockpiling, preparing, correspondence inside the

vehicle and with the climate between different elements, and the control

of self-governing shrewd vehicles are key viewpoints that include

appropriate instruments. Note that these valuable choices are assumed

the premise of huge information examination by AIV (back-end, large

information implies information produced by vehicle correspondence or

web associated objects, permitted in vehicles) [68,69]. Notice that the

main capacity of self-ruling wise vehicle correspondence is to make them

insightful, self-sufﬁcient, and so on. On another side, few requirements

are:

One of the vital prerequisites of the self-sufﬁcient savvy vehicle is

street discovery, which is normally accomplished through different

sensors, for example, the on-board camera and LIDAR. These sensors

require various prerequisites; the highlights of both are saddled by

sensor combination strategies.

Vehicle area is likewise one of the basic practical boundaries for the

self-sufﬁcient savvy vehicle and is consequently reﬁned by informa-

tion from various sensors like GPS, gyro, speed sensors, accelerom-

eter, and so on

Numerous sensors are utilized by AIV to connect and deliver huge

amounts of information. This gathered information is prepared with

exceptional and present-day devices to get full utility/foresee valuable

AIV/AIV/AIV (during movement) subtleties. For example, Oliveira et al.

[68] proposed a system to precisely envision the scene (signiﬁcant for

both AIV observation and getting the hang of) utilizing huge scope

polygonal natives from the 3D information gathered by means of a reach

sensor. Notice that the scene will ceaselessly move, a ﬁxed system that

manages unanticipated conditions, for example, a few street/parkway

snags is hard to execute. Utilizing LIDAR information for street location,

Xiao et al. [69] proposed sensor combination procedures. To tackle the

advantages of both LIDAR and camera sensors, the creators utilized a

variation of Conditional Random Field (CRF), known as Hybrid CRF

(HCRF). This model uses an instrument for paired marking where 'way' or

'history' are named. Associated vehicles and astute independent vehicle

advancements are regularly seen as discrete advances, yet they are just

symmetrical to one another [69]. Vehicle-to-Vehicle (V2V) and

Vehicle-to-Infrastructure (V2I) interchanges, for instance permit vehicles

to discuss helpfully with one another and with equipment foundation

gadgets, for example, RSUs to help a wide scope of uses [70–72] to dis-

tinguish/manufacture trafﬁc sees or to give other street administrations

(making driving more secure for others too). Self-ruling vehicles, then

again, can participate and convey and can give major issues, for example,

bogus area/data, protection breaks, maltreatment of certainty, and so on

Hobert et al. considered the executions of helpful AIV and tended to the

conventions utilized for agreeable interchanges. Moreover, corrections to

existing norms, for example, IEEE 802.11p were likewise proposed to

meet the extra correspondence prerequisites of the helpful self-ruling

savvy vehicle, for example, extra vehicle status information, guard the

board, moving, crossing point the executives, agreeable detecting, high

message rate, low start to ﬁnish delay and improved dependability [73].

Figs. 6 and 7 location all components that require to be incorporated

in self-ruling clever vehicles to make them more modern and insightful

(without human contact) while driving. Note that self-governing helpful

driving additionally helps with platooning where self-ruling vehicles

convey, sharing natural subtleties, and secure working moves. The uses

of the independent astute vehicle require solid, convenient and successful

correspondence. In cell organizations, in any case, asset dispersion is an

issue and should be totally overseen by the vehicle. The required infor-

mation isn't restricted to agreeable awareness in independent savvy

vehicle detachment correspondence, yet in addition company the board,

for example, entering and leaving the unit, etc.

Moreover, LTE-driven [72], Visible Light Communication (VLC) [74]

is regularly utilized in the ideal sight-line connected vehicle setting

where transmitters and beneﬁciaries are mounted in the vehicles' head-

lights and tail lights. In remote organizations, VLC utilizes obvious light

for both brightening and information transmission. Nonetheless, VLC is

still in its early stages, and to meet the prerequisites of associated vehicles

and independent wise vehicle applications, effective channel demon-

strating is fundamental. The immense measure of information that the

self-governing astute vehicle produces and cycles additionally inﬂuences

the current accessible organization data transfer capacity. In this way,

A.K. Tyagi, S.U. Aswathy International Journal of Intelligent Networks 2 (2021) 83–102

analysts need to analyse novel procedures as likely work to fulﬁl the

needs of Autonomous Vehicles data transmission. For helpful movement

arranging, self-ruling vehicles frequently connect with people on foot. A

contact system among walkers and self-ruling vehicles, called Eyes on a

vehicle, was proposed by Chang et al. [75]. As he/she goes across the

street and surveys the walker's motivation, the self-governing keen

vehicle visually connects with the passer-by. The independent canny

vehicle settles on the choice whether to stop or go across the street

dependent on the apparent reason. The reason for these tests was to

gather data on client responses to AIV that would eventually help AIV

originators plan better client association frameworks.

Control is another part of the keen independent vehicle that offers

direction along the normal street. Self-ruling vehicle control is a module

that screens and aides the usage of the independent canny vehicle's ac-

tivities in different circumstances and conditions [76]. Likewise, control

additionally alludes to a portion of the equipment level that changes the

aims delivered by different programming modules into conduct per-

formed by the equipment, for example, when the direction is intended for

independent insightful vehicles, the control module must guarantee that

the self-ruling canny vehicle takes the direction during movement and

handles both expected and unexpected circuits. For the self-ruling

shrewd vehicle, horizontal control and longitudinal control, Jo et al.

[49] built up a two-level control framework. A few researchers have

recently considered control as far as deferrals in correspondence and

made it a help system for between vehicular correspondence [77].

Self-ruling shrewd vehicles manage immense amounts of ongoing in-

formation and on account of platooning applications, self-ruling vehicles

must associate progressively with neighbours and the climate, for

example we have to make the vehicle's control framework adequately

successful to settle the unit's activity. Note that contact defer assumes a

basic function in vehicle/detachment correspondence [78]. Remote

systems administration is defenceless against mistakes in correspondence

and unanticipated deferrals. Independent vehicles should likewise have a

powerful control framework to adapt to the transmission delays between

vehicles brought about by remote interchanges. Recall that the

self-governing keen vehicle control framework contrasts from the con-

ventional control framework. All in all, the self-ruling smart vehicle's

focal control framework depends on the setting where versatile control

frameworks are pushed. Consciousness of the foundation, anyway is

signiﬁcant for the control framework to carry on like that. Liu et al. [79]

utilized contact with vehicles in a unit and with the vehicles that are

important for the company dependent on signals. While this work centres

explicitly around associated vehicles, this joint control-correspondence

instrument is able to do permitting self-ruling vehicles to react and

adjust to different street conditions, both in units and in individual

self-ruling vehicles. In synopsis, various methodologies, for example,

helpful correspondence, are accessible to take in the signiﬁcance from

the climate.

8.4. Decision making for autonomous intelligent vehicles

Dynamic is signiﬁcant, however discussing erratic conditions for AIV

or making a vehicle driverless is troublesome and complex. Vehicles that

typically just focus on their nearby climate (called conscience movement

of vehicles) and don't consider (during plan and execution), just consider

their present status, speed, course, and objective, and so on [80]. Notice

that the need for the neighbourhood climate is exceptionally basic with

the improvements in vehicle organizations/self-governing vehicles,

which must be considered in settling on any choice. We covered a few

forecast calculations and methods in the past segment that are sig-

niﬁcant/helpful as an indispensable piece of AIV for PC vision. These

calculations and methods give a high probability of forecasts. Notwith-

standing, in the wake of considering the expectations dependent on

tactile information and input from different modules, an ofﬁcial choice is

made. However, the dynamic issue here is an unusual world (and high

vehicle portability) that inﬂuences the forecast and, at last, the dynamic

cycle. These outcomes are gotten by clamour in tangible information,

ﬂighty conduct, sensor restrictions, and the neighbours' shrouded state

Fig. 6. Component need to be embedded in Autonomous Intelligent Vehicles.

Fig. 7. High-level functional components of an autonomous vehicle.

A.K. Tyagi, S.U. Aswathy International Journal of Intelligent Networks 2 (2021) 83–102

[81]. The self-governing shrewd vehicle framework must acquire

dependable, ﬁne-grained data about (from the neighbours to make an

expectation with the most elevated likelihood. Be that as it may, in

certain circumstances, such neighbourhood data may not be public or

may not be traded. For self-ruling wise vehicle forecast and observation

modules that straightforwardly impact choices, this restriction presents

difﬁcult issues. It is unbelievably hard to mimic human activities in

self-sufﬁcient vehicles; the dynamic cycle turns out to be signiﬁcantly

more troublesome subsequently. Recollect that the issue of dynamic is

multi-dimensional and depends on different components, for example,

the conduct, discernment and expectation of self-governing vehicles,

neighbours, handling of sensor information, alignment of segments, etc.

In Autonomous Vehicles, current dynamic frameworks can be separated

into AI, profound learning, man-made brainpower, multi-political dy-

namic, and Markov dynamic cycles [78,81–83]. The same number of

scholars have endeavoured to give a superior dynamic strategy in AIV,

they have attempted. The development of AIV innovation will be given

by the neighbourhood climate/personality movement of Autonomous

Vehicles, as examined previously.

Also, between vehicle contact can assume an essential function in the

activities of shared group detecting, swarm observation, and neighbours.

The contact hubs (Autonomous Vehicles, normal vehicles, and others, for

example, walkers) in AIV won't just trade information with one another

yet will likewise share mindfulness and driving choices that will improve

the dynamic cycle abilities. Rather than making the individual excursion

of a self-sufﬁcient shrewd vehicle more secure, the choices taken in this

agreeable way would globally affect the encompassing trafﬁc. Related car

innovation [31] will likewise assume a signiﬁcant function sooner rather

than later commercialization of independent vehicles. Associated vehicle

innovation has been completely contemplated and associated vehicle

innovation usage would already be able to be found in very good quality

vehicles today. Note that as correspondence substances, associated ve-

hicles misuse the two vehicles and organizations. What's more, new ap-

plications utilizing cloud administrations have as of late brought about

the joining of associated vehicles with cloud frameworks [84]. In the

previous quite a while, various researchers have tended to vehicular

distributed computing work in to give sway administrations to

savvy/self-ruling vehicles [33]. The self-ruling wise vehicle exploits the

beneﬁts of the cloud from numerous points of view, for example, the

execution of sensor combination calculations, extensive guide creation,

gadget diagnostics, index of history, and other asset hungry machines

and profound learning calculation [14,85]. The correspondence delay

brought about by correspondence between the self-ruling shrewd vehicle

and the RSU as well as cloud foundation, nonetheless, makes this tech-

nique less alluring to the AIV's basic capacities. Note that in AIV, no

postponement from the RSU can be endured by the dynamic module. The

capacity of the cloud framework is subsequently restricted to esteem

added beneﬁts and long haul AIV conduct investigation, while dynamic is

done progressively at AIV locally. Haze registering, which stretches out

the cloud worldview to the edges of the organization, can be utilized

notwithstanding distributed computing to give constant administrations

requiring low deferrals [86]. Barely any creators have set their endeav-

ours into independent vehicles dependent on haze registering.

8.5. Real-world tests of autonomous intelligent vehicles

Until this point, a few true investigations to decide the activity and

proﬁciency of Autonomous Intelligent Vehicles have been performed.

Campbell et al. [20] partook in the Autonomous Vehicles DARPA Grand

Challenge (DGC). Not many other genuine word tests are depicted top to

bottom in Ref. [87]. This certiﬁable test gave Campbell et al. [20] with

more bits of knowledge into the issues that should be defeated should the

AIV innovation become business. Moreover, Endsley et al. [88] explored

the self-governing keen vehicle Tesla Model S for a half year from

different perspectives, for example, assessing comprehension of the

circumstance, transformation to the self-sufﬁcient smart vehicle,

response to erratic street conditions, etc. The examination reasoned that

client mental model creation, trust in self-governing vehicles, ecological

unpredictability and interfaces/plan for the inhabitants incorporate a

portion of the squeezing difﬁculties looked by the independent astute

vehicle industry. Since this exploration is centred around close to home

and individual experience, from a genuine client point of view, it offers

important input. Broggi et al. [89] at Artiﬁcial Vision and Intelligent

Systems Lab (VisLab) have created and tried Brain drive (BRAiVE). A

progression of tests was performed on a self-governing insightful vehicle

constructed locally that voyaged 13,000 km from Italy to Shanghai. The

VisLab model experienced a few obscure conditions during this under-

taking, and the venture designers had an occasion to assess the effec-

tiveness and execution of the model. VisLab's Intercontinental

Autonomous Challenge (VIAC) was alluded to as this undertaking. Af-

terward, in 2013, by testing Autonomous Vehicles on the interstates,

Broggi et al. [46] brought the VIAC experience to another level. Not

exclusively did the undertaking named PROUD ﬁx the issues with VIAC,

however it likewise voyaged faster than VIAC. While the PROUD test

accomplished its focused-on results, it additionally uncovered that it is

imperative to additionally explore driving execution, self-sufﬁcient

vehicle speed and discernment on multi-path streets. Jo et al. [48,49]

planned an AIV without any preparation and performed deﬁnite tests on

the vehicle's model. The discoveries depend on the result of the 2012

South Korean self-ruling keen vehicle rivalry. The design engineering of

the self-sufﬁcient wise vehicle was created and tried by Jo et al. in

different conditions. A huge part which is liable for the typical working of

the common vehicle and the mobility of the AIV is the product design for

car items.

AUTOSTAR [90] is an open standard engineering utilized by

numerous vehicle makers, among other programming structures. In any

case, AUTOSTAR is excessively exorbitant and too convoluted to even

consider incorporating for research ventures. A lighter variation, in

particular AUTOSTAR light, was thusly proposed in Ref. [91]. The lighter

form of AUTOSTAR was utilized by Jo et al. for their self-ruling insightful

vehicle programming. The creators lean toward a circulated way to deal

with the self-governing savvy vehicle engineering over the incorporated

methodology, with far reaching tests created on the independent keen

vehicle, where utilitarian parts of the self-sufﬁcient shrewd vehicle are

gathered into numerous nearby processing units. The inspiration driving

the utilization of dispersed design is to manage the complexities of canny

self-ruling vehicle calculations. In addition to the fact that it increases

productivity, yet it additionally builds execution by parting the ﬁguring

load into numerous nearby calculation units. Notice that in reality situ-

ation, aside from the scholarly world, the car business has likewise taken

activities to connect with possible clients in Autonomous Vehicles tests,

for example, 'Drive Me' is a particular venture dispatched by Volvo,

where the organization proposed to convey around 100 vehicles to cli-

ents in Sweden to assemble buyer information about their everyday

schedules [92]. These tests/challenges produce a lot of ongoing infor-

mation that is at present being utilized to refresh and improve the pre-

sentation of Autonomous Vehicles with new highlights. We required

some successful or current learning methods that included high

arrangement frameworks to investigate/reﬁne this huge measure of in-

formation. Some different scientists have imparted their encounters to

the AIV and recorded critical perspectives on the issues that the

completely self-sufﬁcient astute vehicle actually should be unravelled.

Robotized Analytic Process: The eventual fate of Autonomous Ve-

hicles will be chosen by their wellbeing, vigour, effortless debasement,

safeguard nature, equipment/programming plans, and purchaser fulﬁl-

ment. Human conduct pantomime in Autonomous Vehicles is very

troublesome, and as result, the dynamic cycle turns out to be much

additionally testing. To this end, the dynamic issue is multi-dimensional

and relies upon different components, for example, independent vehicle's

conduct, discernment and forecast, neighbours, sensor information pre-

paring, segments' adjustment, etc. In not-so-distant future, shrewd me-

ters, brilliant apparatuses, environmentally friendly power assets, and

A.K. Tyagi, S.U. Aswathy International Journal of Intelligent Networks 2 (2021) 83–102

energy proﬁcient assets are need of future and require computerized

examination. Consequently, this part gives an approach to future or

future examination bearings towards AIV. What's more, next area with

close this work in a word with including a few helpful and signiﬁcant

comments for future and future analysts for making Autonomous Intel-

ligent Vehicles innovation a reality.

8.5.1. Future of internet of vehicles in the next decade

Vehicle of Everything (VoE) or Internet of Vehicles (IoV) is the future

of transportation or intelligence in vehicles for next generation society.

Now in near future, vehicles will be communicated, a simple picture in

detail has been explained in Fig. 8.

In Fig. 8, the future of Vehicle of Everything (VoE) can be explained

as: (a) An overview of the communication structure of a smart vehicle; (b)

The basic vehicular communication framework of ITS mainly contains

vehicles, road-side units, smart devices, pedestrians, infrastructure,

people, homes, grids as well as ﬁve types of V2X communications such as

Vehicle-To-Road-Side Unit (V2R), Vehicle-To-Infrastructure (V2I),

Vehicle-To-Vehicle (V2V), Vehicle-To-Pedestrians (V2P), Vehicle-To-

Devices (V2D), Vehicle-To-Networks (V2N), Vehicle-To-Grid (V2G),

And Vehicle-to-Home (V2H).

Further, we explain/should address and examine these issues with

respect to autonomous vehicles or internet of vehicles. There are a few

inquiries that require answers from the AIV societies or its related or-

ganization, as:

a) Is the driver answerable for a mishap or the proprietor of the vehicle

in case of an involved vehicle where the "driver " isn't in successful

control? In an abandoned vehicle, then again, who is answerable for

the mishap?

b) Who is to blame and who might the police accuse of a wrongdoing for

a situation where a walker is hit by a vacant vehicle?

c) When carelessness includes the activities of an individual and a psy-

chological angle, will there be an offense, for example, careless

driving?

o If the AV is utilized to carry out a wrongdoing, for example, bank

theft or psychological oppression, who might be subject?

o Would an involved AV be secured by similar laws if the AV was not

under the inhabitant's position, but rather the tenant is affected by

liquor or opiates, or is restless?

o Which court will have ward to manage matters identifying with

AVs, or will it build up a unique court or council to manage the

subtleties of this innovation?

o These cars will be covered by the company, and protection will or

won't be required.

The lawful commitments, and protection status should be clear during

the dispatch of such vehicles over street as indicated by where the AV is "

monitored " or involved. The inquiries examined above are the valid and

vital perspectives on society in moral and money related terms. Before we

can infer focal points, for example, improved versatility, upgraded se-

curity, decreased gridlock and all the more extra time, these issues must

have straightforward arrangements. The activities of society will char-

acterize and evaluate the movement at which the fate of the car business/

transport needs to occur. We contend that all partners should defeat the

issues found in this work before an enormous arrangement of self-

governing vehicles hits the streets. This examination broke down the

authentic setting, late patterns and improvements, and the anticipated

future for public utilization of a canny transportation framework or

completely self-ruling vehicles (with knowledge). All the researchers

who work for AIV are free to help society in a delicate manner (securing

mankind and nature). Moreover, analysts and researchers are welcome to

proceed with their work in the years to come to make AIV a reality. We

trust, consequently, that in the event that we proceed with our

Fig. 8. Future of intelligent transportation system with emerging technologies.

A.K. Tyagi, S.U. Aswathy International Journal of Intelligent Networks 2 (2021) 83–102

100

endeavours towards AIV, AIV will without a doubt before long be a

reality.

9. Conclusion

Self-governing vehicles are a cutting-edge innovation that will affect

our lives with a wide assortment of elements and advancements that

inﬂuence our perspectives on their handiness, regular daily existence

reconciliation, beneﬁts, and obviously, downsides, for example, the

nonattendance of 100 % secure gadgets and programming, obligation

laws and guidelines, protection, protection and security of information,

network issues, and principles. The reasoning for future self-governing

vehicles must focus not just on empowering more noteworthy security,

way of life changes or monetary advantages, yet in addition on

decreasing the inadequacies of existing models of self-sufﬁcient vehicles.

Self-governing vehicles are the future keen vehicles that should be less

driving, ground-breaking and crash-dodging the future's ideal metro-

politan vehicle. By incorporating and executing different impending in-

novations, we need customary vehicles (associated half and half vehicles)

into a self-ruling vehicle. One thing is without a doubt, whatever occurs

later on, independent vehicles will profoundly change the manner in

which we ride as individuals change from the driver's seat to the front

seat. A few vehicles will require a human presence, for example some-

body who will be accountable for the vehicle's administration. During the

AIV over street usage, exacting principles, protection and obligation is-

sues (where the AV is kept an eye on preparing grounds) were corrected.

Notice that when the AIV is kept an eye on open streets, the law turns out

to be less direct (or less complex). During the testing cycle, the subject of

proprietorship is evident, as it will be that of the organization building up

the innovation.

Author contributions

With an original idea, Dr. Amit Kumar Tyagi and Aswathy S U

conceived this work, planned the schemes, and drafted this manuscript.

Declaration of competing interest

The authors note that there is no conﬂict of interest with respect to

the publication of this report.

References

[1] https://spectrum.ieee.org/transportation/self-driving/accelerating-autonomous-

vehicle-technology.

[2] J.L. Peter. https://www.robsonforensic.com/articles/autonomous-vehicles-sensors

-expert/, September 2018.

[3] Con Cronin, Andrew Conway, Joseph Walsh, State-of-Art Review of Autonomous

Intelligent Vehicle, ISSC, June 2019.

[4] A. Lafrance, Our grandmother's driverless car. https://www.theatlantic.com/techn

ology/archive/2016/06/beep-beep/489029/, 2016 [Online].

[5] T. Kanade, C. Thorpe, W. Whittaker, Autonomous land vehicle project at cmu, in:

Proceedings of the 1986 ACM Fourteenth Annual Conference on Computer Science,

CSC ’86, ACM, New York, NY, USA), 1986, pp. 71–80.

[6] J. Schmidhuber, Robot car history. http://people.idsia.ch/&sim;juergen/robotca

rs.html.

[7] David Elliot, Keen Walter, Lei Miao, Recent Advances in Connected and Automated

Vehicles, 6, April 2019, 2.

[8] Mostafa El-Said, Samah Mansour, Vijay Bhuse, DSRC Based Sensor Pooling Protocol

for Connected Vehicles in Future Smart Cities, vol. 140, 2018.

[9] Kayhan Zrar Ghafoor, Kamalrulnizam Abu Bakar, Marwan Aziz Mohammed, Jaime

Lloret, Vehicular Cloud Computing: Trends and Challenges, Nov 2013.

[10] Neetha George, Jeena Thomas, Authenticating Communication of Autonomous

Vehicles with Artiﬁcial Intelligence, IOP CSMSE, Aug 2018.

[11] C. Howden, U.S. department of transportation issues advance notice of proposed

rulemaking to begin implementation of vehicle-to vehicle communications

technology. https://www.nhtsa.gov/pressreleases/us-department-transportation-

issues-advance-noticeproposed-rulemaking-begin, 2014.

[12] https://www.cbinsights.com/research/autonomous-driverless-vehicles-corporati

ons-list/.

[13] N. Lopez, Nvidia announces a ’supercomputer’GPU and deep learning platform for

self-driving cars. https://thenextweb.com/author/napierlopez/#.tnw_G6F0jhzi,

2016.

[14] J. Contreras-Castillo, S. Zeadally, J.A.G. Ibanez, Solving vehicular ad hoc network

challenges with big data solutions, IET Netw. 5 (4) (2016) 81–84.

[15] M. Gross, A planet with two billion cars, Curr. Biol. 26 (8) (2016) R307–R310.

[16] WHO, The road trafﬁc death rate by who region and income level. http://www.wh

o.int/gho/road_safety/en/, 2016.

[17] Rasheed Hussain, Sherali Zeadally, Autonomous Cars: Research Results, Issues, and

Future Challenges”, 21, 2019, 2.

[18] Juan Rosenzweig, Michael Bartl, A Review and Analysis of Literature on

Autonomous Driving E-Journal: Making-Of Innovation, Oct 2015.

[19] Asif Iqbal Mohammad Faisal, Yignitcanlar Tan, Md Kmaruzzaman, Graham Currie,

Understanding autonomous vehicles: a systematic literature review on capability,

impact, planning, policy, Journal of transport and Land use 12 (1) (Jan 2019).

[20] M. Campbell, M. Egerstedt, J.P. How, R.M. Murray, Autonomous driving in urban

environments: approaches, lessons and challenges, Phil. Trans. Roy. Soc. Lond.:

Mathematical, Physical and Engineering Sciences 368 (1928) 4649–4672, 2010.

[21] R. Okuda, Y. Kajiwara, K. Terashima, A survey of technical trend of adas and

autonomous driving, in: Proceedings of Technical Program - 2014 International

Symposium on VLSI Technology, Systems and Application, VLSI-TSA, April 2014,

pp. 1–4.

[22] D.J. Fagnant, K. Kockelman, Preparing a nation for autonomous vehicles:

opportunities, barriers and policy recommendations, Transport. Res. Pol. Pract. 77

(2015) 167–181.

[23] S.A. Bagloee, M. Tavana, M. Asadi, T. Oliver, Autonomous vehicles: challenges,

opportunities, and future implications for transportation policies, Journal of

Modern Transportation 24 (Dec 2016) 284–303.

[24] B. Paden, M. €

AN~

AaÎp, S.Z. Yong, D. Yershov, E. Frazzoli, A survey of motion

planning and control techniques for self-driving urban vehicles, IEEE Transactions

on Intelligent Vehicles 1 (March 2016) 33–55.

[25] H. Abraham, C. Lee, S. Brady, C. Fitzgerald, B. Mehler, B. Reimer, J.F. Coughlin,

White paper: autonomous vehicles, trust, and driving alternatives: a survey of

consumer preferences. Tech. Rep. 2016-6, MIT AgeLab, Massachusetts Institute of

Technology, Cambridge, MA, June 2016.

[26] J. Joy, M. Gerla, Internet of vehicles and autonomous connected car - privacy and

security issues, in: 2017 26th International Conference on Computer

Communication and Networks (ICCCN), July 2017, pp. 1–9.

[27] G. Bresson, Z. Alsayed, L. Yu, S. Glaser, Simultaneous localization and mapping: a

survey of current trends in autonomous driving, IEEE Transactions on Intelligent

Vehicles 2 (Sept 2017) 194–220.

[28] S. Parkinson, P. Ward, K. Wilson, J. Miller, Cyber threats facing autonomous and

connected vehicles: future challenges, IEEE Trans. Intell. Transport. Syst. (99)

(2017) 1–18.

[29] W. Shi, M.B. Alawieh, X. Li, H. Yu, “Algorithm and hardware implementation for

visual perception system in autonomous vehicle: a survey,”Integration, the VLSI

Journal 59 (2017) 148–156.

[30] L.M. Hulse, H. Xie, E.R. Galea, Perceptions of autonomous vehicles: relationships

with road users, risk, gender and age, Saf. Sci. 102 (2018) 1–13.

[31] J.A. Guerrero-ibanez, S. Zeadally, J. Contreras-Castillo, Integration challenges of

intelligent transportation systems with connected vehicle, cloud computing, and

internet of things technologies, IEEE Wireless Communications 22 (December

2015) 122–128.

[32] https://www.nhtsa.gov/technology-innovation/automated-vehicles-safety.

[33] Rushit Dave, Evelyn Sowells-Boone, Kaushik Dutta Roy, Efﬁcient Data Privacy and

Security in Autonomous Cars, JCSA, May 2019.

[34] A.K. Tyagi, S. Kumari, T.F. Fernandez, C. Aravindan, P3 block: privacy preserved,

trusted smart parking allotment for future vehicles of tomorrow, in: O. Gervasi, et

al. (Eds.), Computational Science and its Applications –ICCSA 2020. ICCSA 2020,

Lecture Notes in Computer Science, vol. 12254, Springer, Cham, 2020, https://

doi.org/10.1007/978-3-030-58817-5_56.

[35] A. Mohan Krishna, Amit Kumar Tyagi, S.V.A.V. Prasad, Preserving privacy in future

vehicles of tomorrow”, J. Cardpulm. Rehabil. 7 (19) (2020) 6675–6684, https://

doi.org/10.31838/jcr.07.19.768.

[36] Amit Kumar Tyagi, Meghna Manoj Nair, Sreenath Niladhuri, Ajith Abraham,

Security, privacy research issues in various computing platforms: a survey and the

road ahead, Journal of Information Assurance &Security 15 (1) (2020) 1–16.

[37] A.K. Tyagi, T.F. Fernandez, S. Mishra, S. Kumari, Intelligent automation systems at

the core of industry 4.0, in: A. Abraham, V. Piuri, N. Gandhi, P. Siarry,

A. Kaklauskas, A. Madureira (Eds.), Intelligent Systems Design and Applications.

ISDA 2020, Advances in Intelligent Systems and Computing, vol. 1351, Springer,

Cham, 2021, https://doi.org/10.1007/978-3-030-71187-0_1.

[38] K. Sravanthi, Vijay Kumar Burugari, Amit Tyagi, Preserving Privacy Techniques for

Autonomous Vehicles, 2020, https://doi.org/10.30534/ijeter/2020/48892020,8.

5180-5190.

[39] R. Varsha, et al., Deep Learning Based Blockchain Solution for Preserving Privacy in

Future Vehicles, 1 Jan. 2020, pp. 223–236.

[40] J. Kamau, A. Ahmed, A. Rebeiro-H, H. Kitaoka, H. Okajima, Z.H. Ripon, Demand

responsive mobility as a service, in: 2016 IEEE International Conference on

Systems, Man, and Cybernetics (SMC), Oct 2016, 001741–001746.

[41] M.J. Anderson, K. Nidhi, D.S. Karlyn, P. Sorensen, C. Samaras, O.A. Oluwatola,

Autonomous vehicle technology: a guide for policymakers, in: RAND Corporation,

2016.

[42] H.S. Kim, S.H. Yoon, M.J. Kim, Y.G. Ji, Deriving future user experiences in

autonomous vehicle, in: Adjunct Proceedings of the 7th International Conference on

Automotive User Interfaces and Interactive Vehicular Applications, AutomotiveUI

’15, ACM, New York, NY, USA), 2015, pp. 112–117.

[43] D. Pierce, Tesla summon hints at how the world of self-driving cars will work

[Online], https://www.wired.com/2016/01/tesla-summon/, 2016.

A.K. Tyagi, S.U. Aswathy International Journal of Intelligent Networks 2 (2021) 83–102

101

[44] L. Li, Y. Liu, J. Wang, W. Deng, H. Oh, Human dynamics-based driver model for

autonomous car, IET Intell. Transp. Syst. 10 (8) (2016) 545–554.

[45] Margarita Martinez-Diaz, Francesc Soriguera, “Autonomous Veicles: Theoretical

and Practical Challenges”, Transportation Research Procedia, Jan 2018.

[46] A. Broggi, P. Cerri, S. Debattisti, M.C. Laghi, P. Medici, D. Molinari, M. Panciroli,

A. Prioletti, “Proud, Public Road urban driverless-car test, IEEE Trans. Intell.

Transport. Syst. 16 (Dec 2015) 3508–3519.

[47] Viktorita Ilkova, Ilka Adrian, Legal Aspects of Autonomous Vehicles: on Overview,

ICPC, June 2017.

[48] K. Jo, J. Kim, D. Kim, C. Jang, M. Sunwoo, Development of autonomous car (part i):

distributed system architecture and development process, IEEE Trans. Ind. Electron.

61 (Dec 2014) 7131–7140.

[49] K. Jo, J. Kim, D. Kim, C. Jang, M. Sunwoo, Development of autonomous car (part ii):

a case study on the implementation of an autonomous driving system based on

distributed architecture, IEEE Trans. Ind. Electron. 62 (Aug 2015) 5119–5132.

[50] S. Woo, H.J. Jo, D.H. Lee, A practical wireless attack on the connected car and

security protocol for in-vehicle can, IEEE Trans. Intell. Transport. Syst. 16 (April

2015) 993–1006.

[51] S. Woo, H.J. Jo, I.S. Kim, D.H. Lee, A practical security architecture for in-vehicle

can-fd, IEEE Trans. Intell. Transport. Syst. 17 (Aug 2016) 2248–2261.

[52] B. Groza, S. Murvay, Security solutions for the controller area network: bringing

authentication to in-vehicle networks, IEEE Veh. Technol. Mag. (99) (2018), 1–1.

[53] J. Janai, F. Güney, A. Behl, A. Geiger, Computer vision for autonomous vehicles:

problems, datasets and state-of-the-art, CoRR (2017). https://arxiv.org/abs/1

704.05519.

[54] X. Chen, K. Kundu, Y. Zhu, H. Ma, S. Fidler, R. Urtasun, 3d object proposals using

stereo imagery for accurate object class detection, CoRR (2016). https://arxiv

.org/abs/1608.07711.

[55] A. Gonzalez, D. V~

AaÎzquez, A.M. L~

As¸pez, J. Amores, On-board object detection:

multicue, multimodal, and multiview random forest of local experts, IEEE

Transactions on Cybernetics 47 (Nov 2017) 3980–3990.

[56] X. Chen, K. Kundu, Z. Zhang, H. Ma, S. Fidler, R. Urtasun, Monocular 3d object

detection for autonomous driving,, in: 2016 IEEE Conference on Computer Vision

and Pattern Recognition (CVPR), June 2016, pp. 2147–2156.

[57] M. Cordts, M. Omran, S. Ramos, T. Rehfeld, M. Enzweiler, R. Benenson, U. Franke,

S. Roth, B. Schiele, “The cityscapes dataset for semantic urban scene understanding,

in: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR),

June 2016, pp. 3213–3223.

[58] J. Baek, J. Kim, E. Kim, Fast and efﬁcient pedestrian detection via the cascade

implementation of an additive kernel support vector machine, IEEE Trans. Intell.

Transport. Syst. 18 (April 2017) 902–916.

[59] M. Bilal, Algorithmic optimisation of histogram intersection kernel support vector

machine-based pedestrian detection using low complexity features, IET Comput.

Vis. 11 (5) (2017) 350–357.

[60] H. Hattori, V.N. Boddeti, K. Kitani, T. Kanade, Learning scene-speciﬁc pedestrian

detectors without real data, in: 2015 IEEE Conference on Computer Vision and

Pattern Recognition (CVPR), June 2015, pp. 3819–3827.

[61] P. Sermanet, K. Kavukcuoglu, S. Chintala, Y. Lecun, Pedestrian detection with

unsupervised multi-stage feature learning, in: 2013 IEEE Conference on Computer

Vision and Pattern Recognition, June 2013, pp. 3626–3633.

[62] D. Xu, W. Ouyang, E. Ricci, X. Wang, N. Sebe, Learning cross-modal deep

representations for robust pedestrian detection,, in: 2017 IEEE Conference on

Computer Vision and Pattern Recognition (CVPR), July 2017, pp. 4236–4244.

[63] Khayyam Hamid, Bahman Javadi, Mahdi Jalili, Reza Jazar, Artiﬁcial Intelligence

and IoT for Autonomous Vehicles, NAEA, Jan 2020.

[64] Y. Tian, K. Pei, S. Jana, B. Ray, Deeptest: automated testing of deep-neural-network-

driven autonomous cars, CoRR abs/1708 (2017), 08559.

[65] C. Chen, A. Seff, A. Kornhauser, J. Xiao, Deepdriving: learning affordance for direct

perception in autonomous driving,, in: 2015 IEEE International Conference on

Computer Vision (ICCV), Dec 2015, pp. 2722–2730.

[66] A. Laddha, M.K. Kocamaz, L.E. Navarro-Serment, M. Hebert, Map-supervised road

detection,, in: 2016 IEEE Intelligent Vehicles Symposium (IV), June 2016,

pp. 118–123.

[67] A. Dairi, F. Harrou, M. Senouci, Y. Sun, Unsupervised obstacle detection in driving

environments using deep-learning-based stereovision, Robot. Autonom. Syst. 100

(2018) 287–301.

[68] M. Oliveira, V. Santos, A.D. Sappa, P. Dias, A.P. Moreira, Incremental scenario

representations for autonomous driving using geometric polygonal primitives,

Robot. Autonom. Syst. 83 (2016) 312–325.

[69] L. Xiao, R. Wang, B. Dai, Y. Fang, D. Liu, T. Wu, Hybrid conditional random ﬁeld-

based camera-lidar fusion for road detection, Inf. Sci. 432 (2018) 543–558.

[70] R. Hussain, F. Abbas, J. Son, S. Kim, H. Oh, Using public buses as mobile gateways

in vehicular clouds, in: 2014 IEEE International Conference on Consumer

Electronics (ICCE), Jan 2014, pp. 175–176.

[71] H. Peng, D. Li, K. Abboud, H. Zhou, H. Zhao, W. Zhuang, X. Shen, Performance

analysis of ieee 802.11p dcf for multiplatooning communications with autonomous

vehicles, IEEE Trans. Veh. Technol. 66 (March 2017) 2485–2498.

[72] H. Peng, D. Li, Q. Ye, K. Abboud, H. Zhao, W. Zhuang, X. Shen, Resource allocation

for cellular-based inter-vehicle communications in autonomous multiplatoons, IEEE

Trans. Veh. Technol. 66 (Dec 2017) 11249–11263.

[73] L. Hobert, A. Festag, I. Llatser, L. Altomare, F. Visintainer, A. Kovacs, Enhancements

of v2x communication in support of cooperative autonomous driving, IEEE

Commun. Mag. 53 (Dec 2015) 64–70.

[74] F. Hussain, H. Farahneh, X. Fernando, A. Ferworn, Vlc enabled foglets assisted road

asset reporting, in: 2017 IEEE 85th Vehicular Technology Conference (VTC Spring),

June 2017, pp. 1–6.

[75] C.-M. Chang, K. Toda, D. Sakamoto, T. Igarashi, Eyes on a car: an interface design

for communication between an autonomous car and a pedestrian, in: Proceedings of

the 9th International Conference on Automotive User Interfaces and Interactive

Vehicular Applications, AutomotiveUI ’17, ACM, New York, NY, USA), 2017,

pp. 65–73.

[76] S.D. Pendleton, H. Andersen, X. Du, X. Shen, M. Meghjani, Y.H. Eng, D. Rus,

M.H. Ang, Perception, planning, control, and coordination for autonomous vehicles,

Machines 5 (1) (2017).

[77] T. Zeng, O. Semiari, W. Saad, M. Bennis, Joint communication and control for

wireless autonomous vehicular platoon systems, CoRR (2018). https://arxiv

.org/abs/1804.05290.

[78] M. Fu, W. Song, Y. Yi, M. Wang, Path planning and decision making for autonomous

vehicle in urban environment, in: 2015 IEEE 18th International Conference on

Intelligent Transportation Systems, Sept 2015, pp. 686–692.

[79] J. Liu, S. Zhang, W. Sun, Y. Shi, In-vehicle network attacks and countermeasures:

challenges and future directions, IEEE Network 31 (5) (2017) 50–58.

[80] Y.W. Seo, R. Rajkumar, “Tracking and estimation of ego-vehicle’s state for lateral

localization, in: 17th International IEEE Conference on Intelligent Transportation

Systems (ITSC), Oct 2014, pp. 1251–1257.

[81] C. Hubmann, M. Becker, D. Althoff, D. Lenz, C. Stiller, Decision making for

autonomous driving considering interaction and uncertain prediction of

surrounding vehicles, in: 2017 IEEE Intelligent Vehicles Symposium (IV), June

2017, pp. 1671–1678.

[82] L. Claussmann, M. Revilloud, S. Glaser, D. Gruyer, “A study on al-based approaches

for high-level decision making in highway autonomous driving,”in 2017, IEEE Int.

Conf. Syst. Man Cybern. (Oct 2017) 3671–3676.

[83] W. Liu, S.W. Kim, S. Pendleton, M.H. Ang, Situation-aware decision making for

autonomous driving on urban road using online pomdp, in: 2015 IEEE Intelligent

Vehicles Symposium (IV), June 2015, pp. 1126–1133.

[84] Daniel Tokody, Imre Janos Mezei, Gyorgy Schuster, Overview of Autonomous

Intelligent Vehicle Systems”, 2017.

[85] S. Liu, J. Tang, C. Wang, Q. Wang, J.L. Gaudiot, A uniﬁed cloud platform for

autonomous driving, Computer 50 (December 2017) 42–49.

[86] M. Aazam, S. Zeadally, K. Harras, Fog computing architecture, evaluation, and

future research directions, IEEE Commun. Mag. 56 (2018) 46–52.

[87] Joel Pereira, Cristiano Premebida, Alireza Asvadi, Francesco Maria Cannata, Test

and evaluation of connected and autonomous vehicles in real-world scenarios, in:

IEEE Intelligent Vehicle Symposium, June 2019.

[88] M.R. Endsley, “Autonomous driving systems: a preliminary naturalistic study of the

tesla model's, Journal of Cognitive Engineering and Decision Making 11 (3) (2017)

225–238.

[89] A. Broggi, M. Buzzoni, S. Debattisti, P. Grisleri, M.C. Laghi, P. Medici, P. Versari,

Extensive tests of autonomous driving technologies, IEEE Trans. Intell. Transport.

Syst. 14 (Sept 2013) 1403–1415.

[90] S. Bunzel, “Autosar –the standardized software architecture, Informatik-Spektrum

34 (Feb 2011) 79–83.

[91] A. Monot, N. Navet, B. Bavoux, F. Simonot-Lion, Multisource software on multicore

automotive ecus: combining runnable sequencing with task scheduling, IEEE Trans.

Ind. Electron. 59 (Oct 2012) 3934–3942.

[92] T. Victor, M. Rothoff, E. Coelingh, A. €

Odblom, K. Burgdorf, When autonomous

vehicles are introduced on a larger scale. In the Road Transport System: the Drive

Me Project, Springer International Publishing, Cham, 2017, pp. 541–546.

A.K. Tyagi, S.U. Aswathy International Journal of Intelligent Networks 2 (2021) 83–102

102

Digital supply chain: literature review of seven related technologies

Article

Full-text available

Apr 2024

This paper systematically reviews literature related with digital supply chains (DSC) and investigates the application status and development trend of different digital technologies in supply chain management. The review is conducted from the perspective of seven key digital supply chain technologies, i.e. Internet of Things (IoT) & Radio Frequency Identification (RFID), 5th Generation Mobile Communication Technology (5G), 3D Printing, Big data (BD), Blockchain, Digital Twins (DT), and Intelligent autonomous vehicles (IAVs). It highlights the main limitations and opportunities of the various DSC technologies, provides an overview of prior studies, and identifies knowledge gaps by outlining the advantages, weaknesses and restrictions of individual technology. The paper also aims at providing a development framework as a roadmap for the match of different digital technologies with different strategic goals.

Intelligent Control of Electric Vehicle Drives using Swarm Robotics

Article

Full-text available

Apr 2024

This study investigates the incorporation of swarm robotics into the control mechanism of electric vehicles (EVs), introducing an innovative intelligent control framework that utilizes the concepts of decentralized decision-making. The research entails a methodical inquiry that encompasses the design of system architecture, the creation of a model for swarm robotics, the modeling of electric vehicle drive, the integration of swarm robotics with EV control, the development of algorithms for intelligent control, and the execution of real-world tests. The fleet of electric cars, propelled by a collective of independent robotic entities, displayed remarkable flexibility in adjusting to fluctuating surroundings. Findings demonstrated disparities in operating duration, distance traversed, mean speed, and energy expenditure during several iterations, highlighting the system’s adeptness in promptly reacting to instantaneous inputs. Significantly, the swarm-propelled electric cars successfully attained varied operating durations, showcasing the system’s adaptability in accommodating environmental dynamics. The swarm-driven system demonstrated its navigation effectiveness by effectively covering various distances, highlighting its versatility and extensive coverage capabilities. The system’s ability to effectively balance energy economy and performance is shown by the collective regulation of average velocity. The energy consumption study demonstrated the system’s efficacy in optimizing energy use, with certain experiments showing significant savings. Percentage change studies have yielded valuable insights into the comparative enhancements or difficulties seen in each indicator, so illustrating the influence of decentralized decision-making on operational results. This study is a valuable contribution to the ever-changing field of intelligent transportation systems, providing insight into the immense potential of swarm-driven electric cars to completely transform sustainable and adaptable transportation. The results highlight the remarkable flexibility and optimization skills of swarm robotics in the management of electric vehicles, paving the way for future advancements in the quest for intelligent, energyefficient, and dynamically responsive transportation solutions.

Load Energy Control of Autonomous Connected Vehicles Based on ML Approach

Conference Paper

May 2024

Evaluating Autonomous Vehicles’ Impact on Emergency Evacuation Clearance Time: A Comparative Study

Conference Paper

Jun 2024

Explainable AI-Based Semantic Object Detection for Autonomous Vehicles

Chapter

Apr 2024

A goal is to facilitate the recognition and segmentation of the route driven by autonomous vehicles through the use of machine learning (ML) models. The pixel-wise road detection task, the semantic segmentation architectures underwent training and comparison. By using XAI, the authors are able to interpret and read the predictions generated by these abstract models. They generated arguments for the recommended segmentation model for autonomous vehicle road detection using a range of XAI approaches. Supervised learning is enabled by KNN, Decision Tree, and Random Forest, which are the current algorithms used for comparison. On the other hand, the newly built K means clustering function best when paired for image processing since they are good with images. To display the results of computations for the evaluation parameters of each algorithm, including accuracy, sensitivity, recall and precision, tables, and the necessary features from the evaluation matrices are utilised. The k-means clustering system for explainable AI-based semantic object detection in automated cars achieves 94.58% accuracy on both train and test sets.

Computer Vision Applications in Intelligent Electric Vehicle Charging Infrastructure

Article

Full-text available

Mar 2024

The study examines the use of computer vision technologies into intelligent electric vehicle (EV) charging infrastructure. The objective is to increase station capabilities, maximize resource usage, and enhance user experiences. An examination of the data from charging stations indicates that there are differences in their capacities and capabilities. Certain stations can handle a greater number of cars at the same time because they have higher power outputs and numerous charging connections. The vehicle identification data illustrates the efficacy of computer vision in precisely recognizing various electric vehicle types, hence optimizing authentication procedures for efficient charging. An analysis of charging session data reveals variations in energy use and durations across sessions, underscoring the impact of charging practices on the utilization of charging stations. An examination of use reveals discrepancies in the number of sessions and energy usage among stations, highlighting the need for adaptive infrastructure. Percentage change study management solutions for demonstrates the fluctuating patterns of resource usage, emphasizing the need for flexible resource allocation techniques. The results emphasize the significant impact that computer vision may have on improving the efficiency and flexibility of electric vehicle charging infrastructure. The research highlights the significance of optimizing the allocation of resources, improving algorithms for various contexts, and applying adaptive solutions for optimal management of charging infrastructure. In essence, the study helps to further our knowledge of how computer vision contributes to the development of intelligent EV charging systems. It provides valuable insights into improving the efficiency of infrastructure and enriching user experiences in the field of electric mobility.

A Conceptual Framework For Acceptance of Autonomous Vehicle In Malaysia International Journal of Management, Finance and Accounting

Article

Full-text available

Jan 2024

Mobility is evolving globally. Automated vehicle technology is consistently advancing with the development of Artificial Intelligence and information communication technology. The Autonomous Vehicle is acknowledged with the benefits of reducing traffic fatalities, reduced emissions and convenience. If autonomous vehicles are widely adopted, the Sustainable Development Goals could be achieved. Various studies have been conducted to investigate the psychological factors (internal) as well as assess the efforts of institutions (external) in promoting the adoption of autonomous vehicles. Nonetheless, there is very few studies examined the impacts of internal and external factors on the acceptance of autonomous vehicles simultaneously. Therefore, this study is taken to close the gap in understanding the acceptance of autonomous vehicles in Malaysia by integrating the internal factors and external factors in a model. By reviewing the past studies, a conceptual framework which can offer a comprehensive insight to the policymakers and car makers from the public's perspective is proposed. Implications from this study can serve as a basis for prioritising the budget resources and development guidelines for the successful implementation of autonomous vehicles in Malaysia.

Security and Privacy Issues of AI in Autonomous Vehicles

Chapter

Feb 2024

Need of 21st Century: A Sustainable Environment Together With a Smart Environment

Chapter

Dec 2023

In this 21st century, the need for a sustainable environment is becoming increasingly essential. This work discusses the importance of adopting sustainable practices and implementing smart technologies to create a useful and eco-friendly living environment. This chapter explains the important need for sustainable practices to mitigate these challenges and ensure a livable future. Smart environment concepts and technologies are then discussed as a means to achieve sustainability. This work discusses the integration of advanced technologies, such as the internet of things (IoT), artificial intelligence (AI), and data analytics, into various aspects of environmental management. Examples include smart energy grids, intelligent waste management systems, and eco-friendly transportation solutions. These technologies enable efficient resource allocation, real-time monitoring, and data-driven decision-making, fostering a sustainable and resilient environment. This work discusses the benefits of adopting a smart environment approach.

Path planning algorithms in the autonomous driving system: A comprehensive review

Article

Apr 2024
ROBOT AUTON SYST

Intelligent Automation Systems at the Core of Industry 4.0

Chapter

Full-text available

Jun 2021

Today’s in 21st century, we require Digital Transformation everywhere and want to make human life easier and longer to live. Digital Transformation cannot be accomplished by companies/ industries without the use of artificial intelligence (AI, i.e., analytics process) and Internet of Things (IoTs) together. AI and IoTs are the necessity of next decade and of many nations. On another side, some other technology like Blockchain technology and edge computing make the integration these technologies simple and faster. In near future, Digital Transformation will require more than one technology, i.e., integration of technologies will be ion trend. The word 'Intelligent Automation,' which is essentially the automation of the processes of the business (including general corporate-level processes using BPM and unique task-level processes using RPA), is therefore assisted by Artificial Intelligence’s analytics and decisions. This work discusses about Intelligent Automation, its internal structure, evolution and importance (with future work) in many useful applications (for Industry 4.0). In last, Intelligent Automation Systems has been explained for e-healthcare applications and give a perspective “How it can change Healthcare Industry and can save millions of lives.

Deep learning based Blockchain solution for preserving privacy in future vehicles

Article

Full-text available

Dec 2020

The Internet of Things (smart things) is used in many sectors and applications due to recent technological advances. One of such application is in the transportation system, which is of primary use for the users to move from one place to another place. The smart devices which were embedded in vehicles are useful for the passengers to solve his/her query, wherein future vehicles will be fully automated to the advanced stage, i.e. future cars with driverless feature. These autonomous cars will help people a lot to reduce their time and increases their productivity in their respective (associated) business. In today’s generation and in the near future, privacy preserving and trust will be a major concern among users and autonomous vehicles and hence, this paper will be able to provide clarity for the same. Many attempts in previous decade have provided many efficient mechanisms, but they all work only with vehicles along with a driver. However, these mechanisms are not valid and useful for future vehicles. In this paper, we will use deep learning techniques for building trust using recommender systems and Blockchain technology for privacy preserving. We also maintain a certain level of trust via maintaining the highest level of privacy among users living in a particular environment. In this research, we developed a framework that could offer maximum trust or reliable communication to users over the road network. With this, we also preserve privacy of users during traveling, i.e., without revealing identity of respective users from Trusted Third Parties or even Location Based Service in reaching a destination. Thus, Deep Learning based Blockchain Solution (DLBS) is illustrated for providing an efficient recommendation system.

P3 Block: Privacy Preserved, Trusted Smart Parking Allotment for Future Vehicles of Tomorrow

Chapter

Full-text available

Sep 2020

As the urbanization is advancing day by day and demand of vehicles is being increased (or cheaper cost of vehicles), it made increased number of vehicles over the road, which faces several issues like traffic accidents, finding a parking spot, etc. Finding a car park slot has become one of the residents’ key pain points. The key reason behind this is the restricted provision of parking lots and the struggle to spot vacant parking space during peak times. To overcome this restriction, one solution is to create smart allocation of free spaces that are easier to locate and use (i.e., trusted and privacy preserved for user or driver). In this work, we provide privacy preserved parking allotment (as a smart solution) to users (or vehicle users) who are willing to pay more charges when they require parking slot near to their destination (or urgent need). In this work (called as P3 Block: Privacy Preserving parking using Blockchain 2.0), one useful and different approach we used as “Incentive based smart parking”. People will get cashback or reward for alerting or spreading genuine information among users. Also, miner will be rewarded for verifying users. Incentive feature will attract more users, also more trust and reputation of specific service provider matter a lot in providing efficient parking slot to vehicle user. Today world is moving towards automation, so automation will be available in all possible applications like manufacturing, finance, retail, food supply management, transportation, home appliances, etc.

Preserving Privacy Techniques for Autonomous Vehicles

Article

Full-text available

Sep 2020

In this automated world, there will be increase in population, more city development and less space for parking results in parking related problems. In our day-to-day life vehicle became an important resource. To go anyplace we require a vehicle, when we are using vehicle, parking will be the big problem. Due to insufficient parking lots and no knowledge about the parking lots (i.e., do not the parking lots places). To park a vehicle manually we have to wait in line to take the token for parking which is a time-consuming process. There is a need to find a secured, smart, efficient system for finding nearest unoccupied parking lots, to reserve parking lot, guidance to navigate to parking lot, to negotiate parking fee and finally to secure the vehicle in parking places. Smart parking system is introduced by Intelligent Transport Systems (ITS). This article gives an overview on different smart parking services like reserving parking lots, navigating to parking lots, finding nearest parking lots, how to provide security to vehicle. It shows the cons and pros of every smart parking system.

Test and Evaluation of Connected and Autonomous Vehicles in Real-world Scenarios

Conference Paper

Full-text available

Jun 2019

Efficient Data Privacy and Security in Autonomous Cars

Article

Full-text available

May 2019

Understanding autonomous vehicles: A systematic literature review on capability, impact, planning and policy

Article

Full-text available

Jan 2019

Advancement in automated driving technology has created opportunities for smart urban mobility. Automated vehicles are now a popular topic with the rise of the smart city agenda. However, legislators, urban administrators, policymakers, and planners are unprepared to deal with the possible disruption of autonomous vehicles, which potentially could replace conventional transport. There is a lack of knowledge on how the new capabilities will disrupt and which policy strategies are needed to address such disruption. This paper aims to determine where we are, where we are headed, what the likely impacts of a wider uptake could be, and what needs to be done to generate desired smart urban mobility outcomes. The methodology includes a systematic review of the existing evidence base to understand capability, impact, planning, and policy issues associated with autonomous vehicles. The review reveals the trajectories of technological development, disruptive effects caused by such development, strategies to address the disruptions, and possible gaps in the literature. The paper develops a framework outlining the inter-links among driving forces, uptake factors, impacts and possible interventions. It concludes by advocating the necessity of preparing our cities for autonomous vehicles, although a wider uptake may take quite some time. © 2019 Asif Faisal, Tan Yigitcanlar, Md Kamruzzaman & Graham Currie.

Autonomous vehicles: theoretical and practical challenges

Article

Full-text available

Jan 2018

Autonomous driving is expected to revolutionize road traffic attenuating current externalities, especially accidents and congestion. Carmakers, researchers and administrations have been working on autonomous drivingfor years and significant progress has been made. However, the doubts and challenges to overcome are still huge, as the implementation of an autonomous driving environment encompasses not only complex automotive technology, but also human behavior,ethics,traffic management strategies, policies, liability, etc.As a result, carmakers do not expect to commercially launch fully driverlessvehicles in the short-term. From the technical perspective, theunequivocal detection of obstacles at high speeds and long distances is one of the greatest difficulties to face. Regarding traffic management strategies, all approaches share the vision that vehicles should behave cooperatively. General V2V cooperation and platooning are options being discussed, both with multiple variants. Various strategies, built from different standpoints, are being designedand validated using simulation. Besides, legal issues have already been arisen in the context of highly-automated driving. They rangefrom the need for special driving licenses to much more intricate topics like liabilityin the event of an accident or privacy issues. All these legal and ethical concerns could hinder the spread of autonomous vehicles once technologically feasible.This paper provides an overview of the current state of the art in the key aspects of autonomous driving. Based on the information received in situ from top research centersin the field and on a literature review, authors highlight the most important advances and findings reached so far, discuss different approaches regarding autonomous traffic and propose a framework for future research.

Computer Vision for Autonomous Vehicles: Problems, Datasets and State of the Art

Book

Jan 2020

Recent years have witnessed enormous progress in AI-related fields such as computer vision, machine learning, and autonomous vehicles. As with any rapidly growing field, it becomes increasingly difficult to stay up-to-date or enter the field as a beginner. While several survey papers on particular sub-problems have appeared, no comprehensive survey on problems, datasets, and methods in computer vision for autonomous vehicles has been published. This monograph attempts to narrow this gap by providing a survey on the state-of-the-art datasets and techniques. Our survey includes both the historically most relevant literature as well as the current state of the art on several specific topics, including recognition, reconstruction, motion estimation, tracking, scene understanding, and end-to-end learning for autonomous driving. Towards this goal, we analyze the performance of the state of the art on several challenging benchmarking datasets, including KITTI, MOT, and Cityscapes. Besides, we discuss open problems and current research challenges. To ease accessibility and accommodate missing references, we also provide a website that allows navigating topics as well as methods and provides additional information.

DeepTest: automated testing of deep-neural-network-driven autonomous cars

Conference Paper

May 2018

Recent advances in Deep Neural Networks (DNNs) have led to the development of DNN-driven autonomous cars that, using sensors like camera, LiDAR, etc., can drive without any human intervention. Most major manufacturers including Tesla, GM, Ford, BMW, and Waymo/Google are working on building and testing different types of autonomous vehicles. The lawmakers of several US states including California, Texas, and New York have passed new legislation to fast-track the process of testing and deployment of autonomous vehicles on their roads. However, despite their spectacular progress, DNNs, just like traditional software, often demonstrate incorrect or unexpected corner-case behaviors that can lead to potentially fatal collisions. Several such real-world accidents involving autonomous cars have already happened including one which resulted in a fatality. Most existing testing techniques for DNN-driven vehicles are heavily dependent on the manual collection of test data under different driving conditions which become prohibitively expensive as the number of test conditions increases. In this paper, we design, implement, and evaluate DeepTest, a systematic testing tool for automatically detecting erroneous behaviors of DNN-driven vehicles that can potentially lead to fatal crashes. First, our tool is designed to automatically generated test cases leveraging real-world changes in driving conditions like rain, fog, lighting conditions, etc. DeepTest systematically explore different parts of the DNN logic by generating test inputs that maximize the numbers of activated neurons. DeepTest found thousands of erroneous behaviors under different realistic driving conditions (e.g., blurring, rain, fog, etc.) many of which lead to potentially fatal crashes in three top performing DNNs in the Udacity self-driving car challenge.

Autonomous Intelligent Vehicles (AIV): Research statements, open issues, challenges and road for future

Abstract and Figures

Recommended publications

Exploration of issues, challenges and latest developments in autonomous cars

Autonomous Cars: Research Results, Issues and Future Challenges

An Analytical Study on Challenges and Road map to Autonomous Vehicles in Indian Context

A Short Review of Innovation in Autonomous Car in Combination with Mechanical and Electronics