Solar Powered 4-Wheel Drive Autonomous Seed Sowing Robot for Rough Terrain

Abstract

In the era of fourth industrial revolution, automation is a growing trend in almost every industry. The innovation in agriculture equipment is one of the major phase for civilized life, and the development of agricultural tools is a fundamental need towards the improvement of agriculture. Farmers use the same traditional methods and equipment for all ages, for example: seeds, spraying, weeding, etc., which have problems such as slow growth rate, irrigation, fertilization, crop monitoring of large areas. An autonomous, low-maintenance and portable robot can serve this purpose more accurately and efficiently with much better performance output. This article introduces the proposed model and design of a solar powered 4-wheel drive robot which works to plant seeds while cultivating the soil in a multi terrain surface. The robot avoids human effort ranging from the field path following to uniformly sowing of seeds at equal distance intervals using field area constraints prescribed by the farmer. This article presents the step by steps designing a robot and the parameters to be considered before creating a prototype. This robot increases the efficiency of seeding by exact measurement and distance, flattening of a surface, spraying of initial water, and also reduces random spreading of seeds problems.

Full text

Solar Powered 4-Wheel Drive Autonomous
Seed Sowing Robot or Rough Terrain
Fawad NASEERa,1, Akhtar RASOOL b, M. Zia QAMMAR b and Rafay SHAHOOD c
a Head of Computer Science Department, Beaconhouse International College, Pakistan
b CS Department Government College University Faisalabad, Pakistan
c CS Department, Beaconhouse International College, Faisalabad, Pakistan
Abstract. In the era of fourth industrial revolution, automation is a growing trend
in almost every industry. The innovation in agriculture equipment is one of the major
phase for civilized life, and the development of agricultural tools is a fundamental
need towards the improvement of agriculture. Farmers use the same traditional
methods and equipment for all ages, for example: seeds, spraying, weeding, etc.,
which have problems such as slow growth rate, irrigation, fertilization, crop
monitoring of large areas. An autonomous, low-maintenance and portable robot can
serve this purpose more accurately and efficiently with much better performance
output. This article introduces the proposed model and design of a solar powered 4-
wheel drive robot which works to plant seeds while cultivating the soil in a multi
terrain surface. The robot avoids human effort ranging from the field path following
to uniformly sowing of seeds at equal distance intervals using field area constraints
prescribed by the farmer. This article presents the step by steps designing a robot
and the parameters to be considered before creating a prototype. This robot increases
the efficiency of seeding by exact measurement and distance, flattening of a surface,
spraying of initial water, and also reduces random spreading of seeds problems.
Keywords. 4-Wheel Robot, Intelligent Farming, Arduino, Seed Sowing, Solar-
powered, Agriculture, 4IR.
1. Introduction
World population is expected to grow about more than 8 billion by 2035, which will
accelerate the global food consumption. To increase yields, the entire planting process
needs to be systematically improved to maximize the growing capability to overcome
this scenario. Agriculture is adapting automation in agricultural products and field
cultivation methods. Subjective processing has become the most practical technology in
agricultural services as it allows understanding, studying and responding to different
circumstances to improve efficiency. Proximity sensor, humidity sensor and image
sensor are different technologies which are mainly used as a judicious combination of
information. An example of such high resolution data is soil analysis. Proximity
detection requires the sensors to be in contact with the ground or very close. It helps
characterize the soil below the surface at a specific location in a field [1]. Generally
experienced farmers are familiar with traditional multi-row planting equipment.
1 Fawad Naseer, Head of Computer Science Department, Beaconhouse International College,
Recognized Teaching Centre of University of London, 54- West Canal Road, Faisalabad, Pakistan; E-mail:
fawad.naseer@bic.edu.pk
f
Intelligent Environments 2021
E. Bashir and M. Luštrek (Eds.)
© 2021 The authors and IOS Press.
This article is published online with Open Access by IOS Press and distributed under the terms
of the Creative Commons Attribution Non-Commercial License 4.0 (CC BY-NC 4.0).
doi:10.3233/AISE210104
255

Farmers have following restrictions in normal classical sowing methods [2]:
• Inconsistent distribution of seeds in the corps field.
• Due to Inconsistent distribution of seeds, which will result in gaps between
plants.
• Dead seeds due to placement of seeds at incorrect depth of soil.
Figure 1. Classical methods of sowing seeds.
Classical methods of sowing and spraying are available in almost all parts of
the country, as shown in Fig 1, which results in low yields [3]. Further factors, such as
insufficient energy availability and low mechanization rates, also affect crop productivity.
Delays in sowing, harvesting, threshing, and improper nursery preparation also reduce
crop productivity. Therefore, agricultural modernization cannot be avoided to meet
population growth and food needs due to rapid industrialization [3].
Several different agricultural mechanisms have been established to speed up the seed
distribution method in the corps field. The manual planter [4] requires farmers to operate
and wheel rotation is used to operate the seed release mechanism. In contrast, electric
seed growers use diesel engines or electric motors to drive internal mechanisms and are
often mounted on tractors. The main drawback of these systems is their low energy
efficiency as they require a lot of labor or fossil fuels to operate.
2. Related Works
Because there is no effective equipment to support farmers, new technology must be
implemented to empower farmers with the capability to ensure and cope up the required
demand of global need.
In [5], H. Pota R Eaton, J Katupitiya and S D Pathirana conclude that sowing bulks
of seeds is necessary because the number of qualified sowers is greatly reduced. Planting
distance and plant population are important factors in maximizing crop yield. By using
the microcontroller, 8051 is used for communication between input and output devices.
The main disadvantage of this model is that it has only one mechanism.
In [6], authors discussed the mechanisms of automatic tillage, sowing, fertilization
and watering. Automatic sowing and fertilization is carried out using solenoids. The soil
moisture sensor is used for automatic watering applications using Rasbeery pi and the
Internet. The Arduino mega 2560 is used for the robot process and the Rasberry pi is
used for communication with the robot. Internet systems are used by farmers to
communicate with robots to run processes automatically. It reduces the burden on
farmers.
In [7], the development of a GSM-based automatic drip irrigation system was the
main focus. The system uses multiple sensors to get the state of the field and the irrigation
F. Naseer et al. / Solar Powered 4-Wheel Drive Autonomous Seed Sowing Robot256

schedule is based on this data. Microcontroller and App / DTMF are used for control and
communication respectively. The author has developed a system for planning and
scheduling irrigation processes based on real-time information on the spot. Several
irrigation issues have been resolved, including the physical labor of farmers to manage
irrigation, wasted water and wasted time.
The above research articles have helped us to understand the different aspects
induced by research on agricultural robots. The robot designed in the literature search
above has many issues with the robot's movement, sowing, power issues and other
accuracy concerns. These questions are effectively addressed in this task. This work also
sheds light on the future range of robots.
3. Theory of Seed Sowing
Sowing seeds is defined as the process of placing seeds under the soil so that it grows to
become the plant. For comparison, planting is an acquaintance with the placing the
propagules of plants in the soil for growing plants. which can be seedlings, roots, tubers,
leaves or cuttings. Seeds may be sown directly or transplanted. The methods of sowing
are enlisted and detailed as under:
x Drilling
x Dibbling
x Broadcasting
x Sowing with bamboo plough
As all mentioned above techniques to sow seeds in field, we come to know that
conventional sowing has restrictions such as not being able to sow uniformly by hand
and insufficient control of sowing depth, and labor demand is high, which results in
sowing at uneven depths and have a negative effect on the total grown plants.
Figure 2. Seed dropping mechanism from seed container along with infrared based seed counter.
The technique which we have develop is that all seeds are collected in a vertical
bottle which is mounted on a robot and filled by the farmers as required and dropped
using a servo motor to flip the slider and drop seed one by one as shown in Fig. 2. There
is an infrared sensor mounted on this mechanism which counts every seed and dropped
precisely according to the need of a farmer. Our technique to sow a seed in a soil is the
F. Naseer et al. / Solar Powered 4-Wheel Drive Autonomous Seed Sowing Robot 257

drilling mechanism, which is mounted on a robot, it drills the soil and then insert the seed
by using the stated mechanism.
4. Design & Development of a Proposed Robot
There are a lot of fields where Artificial Intelligence is impacting human life with better
performance results along with less human interaction. The proposed design of our robot
is a large scale model to portray working capabilities with better performance including
high accuracy of placing a seed under soil and placing seeds with accurate measured
distance.
4.1. Construction
At initial stage robot mechanical design was developed using computer-aided design
(CAD) and later we built it by using steel pipes, which is shown in Figure 3. The
cylindrical seed boxes have been chosen to pour the seeds in bulk and for that 1.5-liter
PET bottle have been used because they are light, cheap and easy to obtain. The payload
consisting of seed boxes and electronic boxes is assigned a maximum weight of 23 kg.
Figure 3. Steel rectangular pipe based robot chassis frame with four dc motors and tires.
We have designed the size of our robot frame as 20-inch x 30-inch x 13-inch to fit
the payload frame. The frame is made of steel, which provides strong support for tough
terrain, prevents the robot from rolling over, and has the power to effectively control its
movements. Each tire is equipped with four DC motors, enabling the robot to move
dynamically over rugged terrain and control it in a variety of circumstances at any cost.
In the figure 3 shows a drive system consisting of four DC motors connected to all four
buses. The diameter of the sprocket connected between the DC motor and bus is
determined using a formula:
݀ൌ ௣
ୱ୧୬൫ଵ଼଴ ௭
ൗ൯ (1)
Where ݌ is the pith of the sprocket, ݀ is the diameter of the sprocket and ݖ is the
number of teeth on the sprocket. All the variables are decided on the concern of the
required speed of the robot to be smooth and stable over different rough terrain. We can
extract the dimension of the sprocket to be 93.76 mm by using the Eqn. (1) by using the
number of teeth to be 17.
F. Naseer et al. / Solar Powered 4-Wheel Drive Autonomous Seed Sowing Robot258

Once we able to get the dimension of the sprocket, from this we can calculate the
speed of our robot by using Eqn. (2), where ܰ is the rotation per minute of a sprocket
and for this robot we use it to be 200 rpm:
ݒൌగௗே
଺଴ (2)
we can use Eqn. 3 to calculate the required drive power transmitted of all four motors
in a robot, each with a stall current of 10A. By using the dimension and velocity which
we have calculated and maximum torque to be generalized to be 5 Nm as mentioned on
each motor.
ܲൌ
ଶగே்
଺଴ (3)
The block diagram consists of an Arduino microcontroller, which is the main
controller of the entire system along with DC motors, sensors and all helping components
as shown in Fig 4. Solar panels connected to the battery for energy storage, as well as a
charging circuit that supplies + 5 V to the Arduino board. Our battery system provides +
12 V power supply to control the DC motor using the L298 motor driver module.
Figure 4. Blok diagram of complete electronic components
4.2. Electronic Components
The electronic system of robot consists of a microcontroller and drive circuits required
to connect a robot to various mechanical parts as shown in Fig 4. The microcontroller
(MCU) is an Arduino Mega 2560, which is the main controller and brain of a complete
electronic system. This was chosen because of its high processing power and wide range
of peripherals attaching capability, which makes it easy to apply control algorithms.
When the robot is placed on the surface of the field, it intelligently orients around the
field to cover all areas. The edge of the field was detected using the ultrasonic proximity
sensor HC-SR04. This component is used to identify areas that reside on field boundaries
[8]. The MCU uses a motor driver connected to a DC motor to control the movement of
the robot. Each motor driver consists of a H bridge power MOSFET connected to the
MCU.
4.2.1. Arduino Mega 2560
Arduino is an open source physical computing platform based on a simple I / O board
and development environment. It is an Atmega 2560 based microcontroller board, which
F. Naseer et al. / Solar Powered 4-Wheel Drive Autonomous Seed Sowing Robot 259

is used to interact with all sensors and actuators. It keeps all the components well
synchronized for the proper functioning of the robot.
4.2.2. DC Motor Driver L298N
The L298N is a H-bridge driver with high current load. The L298N can supply up to 1A
at voltages between 4.5 and 36. The L298N is designed to drive inductive loads such as
solenoids, relays, DC motors, bipolar stepper motors, and other powerful motors.
application of current or high voltage.
4.2.3. Ultrasonic Sensor HC-SR04
HC-SR04 ultrasonic sensor is used to calculate the distance to an object using sonar
signal. It uses a non-contact ultrasonic sounder to measure the distance to an object and
consists of two ultrasonic transmitters (mainly loudspeakers), a receiver and a control
circuit [9]. The transmitter emits high-frequency ultrasonic waves that are reflected off
nearby solids, and the receiver tracks the reflected echo. This echo is then processed by
the control circuit to calculate the time difference between the transmit and receive
signals [10].
4.2.4. Servo Motor
Servo motors are half-turn actuators used in applications that require a high degree of
swing angle accuracy. This is because the servo motor has a feedback sensor. The
feedback sensor measures the difference between the set and the required angle to ensure
that you reach the end position more accurately.
4.2.5. Power Source
A battery is a source of energy that provides a surge (voltage) of energy to circulate
current in a circuit. It powers the entire circuitry including all the components attached
to the system. Our entire system is powered by a pair of 18650 Li-ion batteries through
the battery management system(BMS) [11]. Sonar sensors have a built-in voltage
regulator; hence it is powered from any DC voltage from 5 volts. Modules are a group
of cells that are electrically connected and packaged in a frame (most commonly known
as a solar panel).
5. Functioning of The Proposed Robot
The working principle of the developed system is described as a flowchart in Fig. 5. The
code is developed to enable the robot to move in all directions which is being controlled
intelligent by the microcontroller Arduino Mega 2560 to automate the whole process of
seeding.
F. Naseer et al. / Solar Powered 4-Wheel Drive Autonomous Seed Sowing Robot260

Figure 5. Flowchart of functioning of proposed robot
The robot starts to move forward and the seeds are planted evenly at an equal
distance. The seeds are stored in a tapered seed bottle, so the seeds can easily enter the
funnel through the attached hose. The seed placement mechanism runs smoothly and
uses servo motors to complete the seeding tasks. It consists of two-stroke sliders used to
control the opening of the hole from which the seeds are poured. After each drop of a
seed, there is an infrared sensor which is used to count the number of seeds passed
through the funnel to the hole. After pouring of pair of seeds, water sprinkle is placed,
its duty is to sprinkle water on that sowed seed. After watering task is to flat the surface
and for that a flap which is attached behind the robot to cover the soil as shown in Fig 6.
Figure 6. (a) Seeding drop mechanism, (b) Water sprinkler, (c) Flapper to flap surface & (d) Seed container.
When the robot deviates from the intended direction of motion due to
undulations in the tillage path, the input provided by the ultrasonic sensor varies. This
aberration is reduced by using the microcontroller which alters the power of the
individual motors as described in Table 1. Thus the controller allows the device to
traverse in the correct direction.
F. Naseer et al. / Solar Powered 4-Wheel Drive Autonomous Seed Sowing Robot 261

Table 1. Motor direction controlling algorithm
Movement Type
Motor A(Left Front)
Motor B(Left Rear)
Motor C(Right Front)
Motor D(Right Rear)
Straight
ON
ON
ON
ON
Left
OFF
OFF
ON
ON
Right
ON
ON
OFF
OFF
Motor A and B are on left side of a robot and Motor C and D are on right hand
side of a robot. Ultrasonic sensors are used to determine the end of the field and, when
detected, rotate the robot in different directions of travel. This allows the robot to plant
seeds in subsequent rows that are zigzag known as tillage rows as shown in Fig 7.
Figure 7. (a) Top view of robot, (b) Front bottom view of robot & (c) Side view of robot.
6. Results and Discussion
In the first series of experiments which was performed to validate the possibility to drill
the accurate required depth to pour a pair of pea seeds to germinate. In this experiment,
the planting testbed replicated the field environment which comprises of soil. The drill
was prepared so that the precisely hole can be achieved at precise depth and pouring of
2 seeds as a pair in that hole. The prepared drill was prepared and dig into soil and waited
for 30 days after successfully sowing of seeds. The moisture in the soil would dissolve
the seeds entirely. All seeds inside a hole successfully germinated into pea cot yledon
which is shown in Fig. .
Figure 8. Seed germination after drilling hole from a robot.
8
F. Naseer et al. / Solar Powered 4-Wheel Drive Autonomous Seed Sowing Robot262

In the second series of experiments, we have tried to sow different types of seeds
and analyze required depth for each seed and the required spacing between sowing of
seeds in field, we also analyze the size of different seeds and come to know that each
seeds with different size required different size of digging hole for sowing, but our robot
at current state is not able to dig a customized hole according to the seed, it ca be a future
work to implement. Table shows the comparison of the types of seeds, their dimension,
required spacing of sowing and the required depth.
Table 2. Seed Analysis and Comparison
Type of seed
Length(mm)
Width (mm)
Diameter(mm)
Spacing(inch)
Depth(inch)
Pea
5.91
5.01
5.18
5-7.5
5
Tomato
6.3
6.9
4.23
6.5-8
5
Sun Flower
9.52
5.12
3.27
6-7
3
Chia
2.11
1.3
0.8
3.5-5
4
Flax
4.64
2.37
1.00
4.5-6
5
Figure 9 shows the graph of approximation of dropping seeds inside the hole
dig by the drill in the different experiments we did before installing that seeding part on
the robot. The graph explains that number of times the dropping of seeds and its accuracy
to drop exactly inside the hole or scattering around the hole.
Figure 9. Experiment result of conducting different times seed dropping in hole.
7. Conclusion and Future Work
The design and development of a robot along with the Working procedure of the
autonomous seed sowing mechanism have been proposed for the sole purpose of which
is to automate the seed sowing process with flattening of the surface and watering over
newly sowed seeds is discussed in this paper. All components except the sensors are
individually designed, and the entire seed mechanism workflow and sequence of
executions are illustrated in the block diagram above. Different models of seed collectors
have also been developed, depending on the type of seed.
For the future purpose, the robot can be equipped with Global Positioning System
(GPS) for more accuracy of the movement of robot and also for future perspective, robot
can be equipped with more powerful processing board which can be used to calculate
intelligent algorithm including machine learning algorithm for better performance of
robot in context of accuracy in navigation around the field.
2
F. Naseer et al. / Solar Powered 4-Wheel Drive Autonomous Seed Sowing Robot 263

8. Acknowledgment
We would like to thank our respected Country Head Ms Raana Sarmad, for the
inspiration and motivation towards creating a research culture in undergraduate students.
References
[1] PJ, S. K., “A brief survey of classification techniques applied Sheela to soil fertility prediction”,
International Conference on Emerging Trends, 2015.
[2] Pratik Hore. a, Goluprasad Gupta. a, Shrikrishn Deshmukh, “Modern seed sowing techniques and
developing technologies,” International Journal of Innovative and Emerging Research in Engineering,
Volume 4, Issue 3, 2017.
[3] D. Ramesh, H.P. Girishkumar, Agricultural Seed Sowing Equipments: A Review International Journal
of Science, Engineering and Technology Research (IJSETR), Volume 3, Issue 7, July 2014
[4] D. Ramesh and H.P. Girishkumar, “Agriculture Seed Sowing Equipments : A Review”, International
Journal of Science, Engineering and Technology Research (IJSETR), vol. 3, Issue 7, July 2014, pp. 1987
- 1992.
[5] H. Pota, R. Eaton, J. Katapriya and S. D. Pathirana, “Agricultural robotics: A streamlined approach to
realization autonomous farming,” in IEEE conference on industrial and information systems, 2007, pp.
85-90.
[6] Gollakota, A. and Srinivas, M.B., (2011), December. “Agribot —a multipurpose agricultural robot”, In
India Conference (INDICON), 2011 Annual IEEE (pp. 1-4). IEEE
[7] Lalwani, A., et al., 2015, December, “A Review: Autonomous Agribot For Smart Farming”, Proceedings
of 46th IRF International Conference, 27th December 2015, Pune, India
[8] J. Borenstein and Y. Koram, “Obstacle Avoidance with Ultrasonic Sensors”, IEEE Journal of Robotics
and Automation, vol. 4, no. 2, April 1988, pp. 213-218.
[9] S. Barai, D. Biswas and B. Sau, "Estimate distance measurement using NodeMCU ESP8266 based on
RSSI technique," 2017 IEEE Conference on Antenna Measurements & Applications (CAMA), Tsukuba,
2017, pp. 170-1
[10] F. Noor, M. Swaied, M. AlMesned and N. AlMuzini, "A Method to Detect Object's Width with Ultrasonic
Sensor," 2018 International Conference on Computing, Electronics & Communications Engineering
(iCCECE), Southend, United Kingdom, 2018, pp. 266-271
[11] Khanna,A;Ranjan,“Solar-powred Android based Speed Control of DC motors through Secure Bluetooth,”
Communication systems and network technologies CSNT 2015 international conference (IEEE
Publication), pp 1244-1249.
[12] P. Amiribavandpour, A. Kapoor, W. Shen and J. Shearer, "Thevmathematical model of 18650 lithium-
ion battery in electric vehicles," 2013 IEEE 8th Conference on Industrial Electronics and Applications
(ICIEA), Melbourne, VIC, Australia, 2013, pp. 1264-1269
F. Naseer et al. / Solar Powered 4-Wheel Drive Autonomous Seed Sowing Robot264