ARGroHBotS: An Affordable and Replicable Ground Homogeneous Robot Swarm Testbed

Abstract

Currently, multi-agent mobile robotic testbeds are expensive, inaccessible, and restricted characteristics. Besides, the remotely accessible testbeds have limitations for specific tasks and limited time of implementation. For these reasons, we designed, produced, and implemented a testbed with excellent qualities for everybody. The ARGroHBotS is an inexpensive, robust, accessible, and flexible testing scenario where anyone can test control and motion theories. ARGroHBotS could be adjusted and set up for specific implementations. Due to the ability to modify both the hardware and the software, it is possible to change both the peripherals and the control technique used. This paper describes the design, operation, and advantages of ARGroHBotS. This system is specifically designed as an excellent possibility to produce an institutional platform where multi-agent and swarm algorithms could be implemented and tested. The used components and technologies allow easy replication and low-cost access, regardless of the global location. ARGroHBotS provides excellent advantages to research groups compared with commercial Robots.

Full text

IFAC PapersOnLine 54-13 (2021) 256–261
ScienceDirect
Available online at www.sciencedirect.com
2405-8963 Copyright © 2021 The Authors. This is an open access article under the CC BY-NC-ND license
.
Peer review under responsibility of International Federation of Automatic Control.
10.1016/j.ifacol.2021.10.455
10.1016/j.ifacol.2021.10.455 2405-8963
Copyright ©
2021 The Authors. This is an open access article under the CC BY-NC-ND license
(
https://creativecommons.org/licenses/by-nc-nd/4.0/
)
ARGroHBotS: An Affordable and
Replicable Ground Homogeneous Robot
Swarm Testbed
Nestor I. Ospina ∗Eduardo Mojica-Nava ∗Luis G Jaimes ∗∗
Juan M. Calder´on ∗∗∗
∗Electrical and Electronic Department, Universidad Nacional de
Colombia, Bogot´a, Colombia (e-mail: {niospinag,
eamojican}@unal.edu.co)
∗∗ Department of Computer Science, Florida Polytechnic University,
Lakeland, FL 33805, (e-mail:ljaimes@floridapoly.edu).
∗∗∗ Bethune-Cookman University, Daytona Beach, FL, USA (e-mail:
calderonj@cookman.edu), Universidad Santo Tom´as, Bogot´a, Colombia
(e-mail: juancalderon@usantotomas.edu.co)
Abstract: Currently, multi-agent mobile robotic testbeds are expensive, inaccessible, and
restricted characteristics. Besides, the remotely accessible testbeds have limitations for spe-
cific tasks and limited time of implementation. For these reasons, we designed, produced,
and implemented a testbed with excellent qualities for everybody. The ARGroHBotS is an
inexpensive, robust, accessible, and flexible testing scenario where anyone can test control and
motion theories. ARGroHBotS could be adjusted and set up for specific implementations. Due
to the ability to modify both the hardware and the software, it is possible to change both the
peripherals and the control technique used. This paper describes the design, operation, and
advantages of ARGroHBotS. This system is specifically designed as an excellent possibility to
produce an institutional platform where multi-agent and swarm algorithms could be imple-
mented and tested. The used components and technologies allow easy replication and low-cost
access, regardless of the global location. ARGroHBotS provides excellent advantages to research
groups compared with commercial Robots.
Keywords: Mechatronic Systems and Robotics, Intelligent Systems and Applications,
Sustainable Design and Control, Testbed, Multi-agent, Autonomous vehicles.
1. INTRODUCTION
In the last few years, the area of scientific research ex-
perienced dramatic growth. Moreover, the most common
way to test the advantage and qualities of new scientific
theories is by simulation software. Nevertheless, simulation
testing is enough to get a general idea of the behavior of
those theories, but no for physical implementation. For
that reason, it is necessary to use a real testing scenario
that warrants safety, the accuracy of data obtained, and
similarity to the focus environment of the research prob-
lem.
There is an area in the robotic field that studies the
behavior and how to stabilize, control, and synchronize
multiple robots for a specific task. Swarm Robotic systems
are inspired by the behavior of social animals such as
ants, bees, birds, and fish, which exemplify how simple
individual actions can generate complex social behaviors.
Swarm robotics tries to imitate these simple individual
actions to create social behaviors with evident character-
istics of robustness, scalability, and flexibility, as explained
by Camazine et al. (1999). As a result, these systems
have many potential application fields such as nano and
micro-robotics, precision agriculture, military robots, au-
tonomous army, search and rescue of victims in disaster
zones, remote sensing, mobility and transportation, and
delivery of last-mile, among others.
Over the last decades, swarm robotics have been re-
searched using different approaches to different behaviors,
such as aggregation, pattern formation, self-assembly, col-
lective exploration, coordinated motion, collective trans-
portation, consensus achievement, and task allocation.
Several research approaches around swarm robotics have
carried out experiments theoretically, using simulation en-
vironments, and in a few cases using physical robotics
platforms. This last case is the least common, given the
complexity of swarm robotics systems and the difficulty
of having physical robotic platforms. Those platforms are
needed to evaluate and verify the hypotheses and theories
of each researches field. Nevertheless, most researchers
evaluate their works through physical robotics platforms,
mostly with homogeneous terrestrial or aerial robots.
Despite the efforts to develop cheaper platforms by re-
search groups, commercial robots are highly used by re-
search projects. Commercial platforms have good advan-
tages, such as short building times or elaborated hardware
for specific tasks. However, these platforms have disadvan-
ARGroHBotS: An Affordable and
Replicable Ground Homogeneous Robot
Swarm Testbed
Nestor I. Ospina ∗Eduardo Mojica-Nava ∗Luis G Jaimes ∗∗
Juan M. Calder´on ∗∗∗
∗Electrical and Electronic Department, Universidad Nacional de
Colombia, Bogot´a, Colombia (e-mail: {niospinag,
eamojican}@unal.edu.co)
∗∗ Department of Computer Science, Florida Polytechnic University,
Lakeland, FL 33805, (e-mail:ljaimes@floridapoly.edu).
∗∗∗ Bethune-Cookman University, Daytona Beach, FL, USA (e-mail:
calderonj@cookman.edu), Universidad Santo Tom´as, Bogot´a, Colombia
(e-mail: juancalderon@usantotomas.edu.co)
Abstract: Currently, multi-agent mobile robotic testbeds are expensive, inaccessible, and
restricted characteristics. Besides, the remotely accessible testbeds have limitations for spe-
cific tasks and limited time of implementation. For these reasons, we designed, produced,
and implemented a testbed with excellent qualities for everybody. The ARGroHBotS is an
inexpensive, robust, accessible, and flexible testing scenario where anyone can test control and
motion theories. ARGroHBotS could be adjusted and set up for specific implementations. Due
to the ability to modify both the hardware and the software, it is possible to change both the
peripherals and the control technique used. This paper describes the design, operation, and
advantages of ARGroHBotS. This system is specifically designed as an excellent possibility to
produce an institutional platform where multi-agent and swarm algorithms could be imple-
mented and tested. The used components and technologies allow easy replication and low-cost
access, regardless of the global location. ARGroHBotS provides excellent advantages to research
groups compared with commercial Robots.
Keywords: Mechatronic Systems and Robotics, Intelligent Systems and Applications,
Sustainable Design and Control, Testbed, Multi-agent, Autonomous vehicles.
1. INTRODUCTION
In the last few years, the area of scientific research ex-
perienced dramatic growth. Moreover, the most common
way to test the advantage and qualities of new scientific
theories is by simulation software. Nevertheless, simulation
testing is enough to get a general idea of the behavior of
those theories, but no for physical implementation. For
that reason, it is necessary to use a real testing scenario
that warrants safety, the accuracy of data obtained, and
similarity to the focus environment of the research prob-
lem.
There is an area in the robotic field that studies the
behavior and how to stabilize, control, and synchronize
multiple robots for a specific task. Swarm Robotic systems
are inspired by the behavior of social animals such as
ants, bees, birds, and fish, which exemplify how simple
individual actions can generate complex social behaviors.
Swarm robotics tries to imitate these simple individual
actions to create social behaviors with evident character-
istics of robustness, scalability, and flexibility, as explained
by Camazine et al. (1999). As a result, these systems
have many potential application fields such as nano and
micro-robotics, precision agriculture, military robots, au-
tonomous army, search and rescue of victims in disaster
zones, remote sensing, mobility and transportation, and
delivery of last-mile, among others.
Over the last decades, swarm robotics have been re-
searched using different approaches to different behaviors,
such as aggregation, pattern formation, self-assembly, col-
lective exploration, coordinated motion, collective trans-
portation, consensus achievement, and task allocation.
Several research approaches around swarm robotics have
carried out experiments theoretically, using simulation en-
vironments, and in a few cases using physical robotics
platforms. This last case is the least common, given the
complexity of swarm robotics systems and the difficulty
of having physical robotic platforms. Those platforms are
needed to evaluate and verify the hypotheses and theories
of each researches field. Nevertheless, most researchers
evaluate their works through physical robotics platforms,
mostly with homogeneous terrestrial or aerial robots.
Despite the efforts to develop cheaper platforms by re-
search groups, commercial robots are highly used by re-
search projects. Commercial platforms have good advan-
tages, such as short building times or elaborated hardware
for specific tasks. However, these platforms have disadvan-
ARGroHBotS: An Affordable and
Replicable Ground Homogeneous Robot
Swarm Testbed
Nestor I. Ospina ∗Eduardo Mojica-Nava ∗Luis G Jaimes ∗∗
Juan M. Calder´on ∗∗∗
∗Electrical and Electronic Department, Universidad Nacional de
Colombia, Bogot´a, Colombia (e-mail: {niospinag,
eamojican}@unal.edu.co)
∗∗ Department of Computer Science, Florida Polytechnic University,
Lakeland, FL 33805, (e-mail:ljaimes@floridapoly.edu).
∗∗∗ Bethune-Cookman University, Daytona Beach, FL, USA (e-mail:
calderonj@cookman.edu), Universidad Santo Tom´as, Bogot´a, Colombia
(e-mail: juancalderon@usantotomas.edu.co)
Abstract: Currently, multi-agent mobile robotic testbeds are expensive, inaccessible, and
restricted characteristics. Besides, the remotely accessible testbeds have limitations for spe-
cific tasks and limited time of implementation. For these reasons, we designed, produced,
and implemented a testbed with excellent qualities for everybody. The ARGroHBotS is an
inexpensive, robust, accessible, and flexible testing scenario where anyone can test control and
motion theories. ARGroHBotS could be adjusted and set up for specific implementations. Due
to the ability to modify both the hardware and the software, it is possible to change both the
peripherals and the control technique used. This paper describes the design, operation, and
advantages of ARGroHBotS. This system is specifically designed as an excellent possibility to
produce an institutional platform where multi-agent and swarm algorithms could be imple-
mented and tested. The used components and technologies allow easy replication and low-cost
access, regardless of the global location. ARGroHBotS provides excellent advantages to research
groups compared with commercial Robots.
Keywords: Mechatronic Systems and Robotics, Intelligent Systems and Applications,
Sustainable Design and Control, Testbed, Multi-agent, Autonomous vehicles.
1. INTRODUCTION
In the last few years, the area of scientific research ex-
perienced dramatic growth. Moreover, the most common
way to test the advantage and qualities of new scientific
theories is by simulation software. Nevertheless, simulation
testing is enough to get a general idea of the behavior of
those theories, but no for physical implementation. For
that reason, it is necessary to use a real testing scenario
that warrants safety, the accuracy of data obtained, and
similarity to the focus environment of the research prob-
lem.
There is an area in the robotic field that studies the
behavior and how to stabilize, control, and synchronize
multiple robots for a specific task. Swarm Robotic systems
are inspired by the behavior of social animals such as
ants, bees, birds, and fish, which exemplify how simple
individual actions can generate complex social behaviors.
Swarm robotics tries to imitate these simple individual
actions to create social behaviors with evident character-
istics of robustness, scalability, and flexibility, as explained
by Camazine et al. (1999). As a result, these systems
have many potential application fields such as nano and
micro-robotics, precision agriculture, military robots, au-
tonomous army, search and rescue of victims in disaster
zones, remote sensing, mobility and transportation, and
delivery of last-mile, among others.
Over the last decades, swarm robotics have been re-
searched using different approaches to different behaviors,
such as aggregation, pattern formation, self-assembly, col-
lective exploration, coordinated motion, collective trans-
portation, consensus achievement, and task allocation.
Several research approaches around swarm robotics have
carried out experiments theoretically, using simulation en-
vironments, and in a few cases using physical robotics
platforms. This last case is the least common, given the
complexity of swarm robotics systems and the difficulty
of having physical robotic platforms. Those platforms are
needed to evaluate and verify the hypotheses and theories
of each researches field. Nevertheless, most researchers
evaluate their works through physical robotics platforms,
mostly with homogeneous terrestrial or aerial robots.
Despite the efforts to develop cheaper platforms by re-
search groups, commercial robots are highly used by re-
search projects. Commercial platforms have good advan-
tages, such as short building times or elaborated hardware
for specific tasks. However, these platforms have disadvan-
ARGroHBotS: An Affordable and
Replicable Ground Homogeneous Robot
Swarm Testbed
Nestor I. Ospina ∗Eduardo Mojica-Nava ∗Luis G Jaimes ∗∗
Juan M. Calder´on ∗∗∗
∗Electrical and Electronic Department, Universidad Nacional de
Colombia, Bogot´a, Colombia (e-mail: {niospinag,
eamojican}@unal.edu.co)
∗∗ Department of Computer Science, Florida Polytechnic University,
Lakeland, FL 33805, (e-mail:ljaimes@floridapoly.edu).
∗∗∗ Bethune-Cookman University, Daytona Beach, FL, USA (e-mail:
calderonj@cookman.edu), Universidad Santo Tom´as, Bogot´a, Colombia
(e-mail: juancalderon@usantotomas.edu.co)
Abstract: Currently, multi-agent mobile robotic testbeds are expensive, inaccessible, and
restricted characteristics. Besides, the remotely accessible testbeds have limitations for spe-
cific tasks and limited time of implementation. For these reasons, we designed, produced,
and implemented a testbed with excellent qualities for everybody. The ARGroHBotS is an
inexpensive, robust, accessible, and flexible testing scenario where anyone can test control and
motion theories. ARGroHBotS could be adjusted and set up for specific implementations. Due
to the ability to modify both the hardware and the software, it is possible to change both the
peripherals and the control technique used. This paper describes the design, operation, and
advantages of ARGroHBotS. This system is specifically designed as an excellent possibility to
produce an institutional platform where multi-agent and swarm algorithms could be imple-
mented and tested. The used components and technologies allow easy replication and low-cost
access, regardless of the global location. ARGroHBotS provides excellent advantages to research
groups compared with commercial Robots.
Keywords: Mechatronic Systems and Robotics, Intelligent Systems and Applications,
Sustainable Design and Control, Testbed, Multi-agent, Autonomous vehicles.
1. INTRODUCTION
In the last few years, the area of scientific research ex-
perienced dramatic growth. Moreover, the most common
way to test the advantage and qualities of new scientific
theories is by simulation software. Nevertheless, simulation
testing is enough to get a general idea of the behavior of
those theories, but no for physical implementation. For
that reason, it is necessary to use a real testing scenario
that warrants safety, the accuracy of data obtained, and
similarity to the focus environment of the research prob-
lem.
There is an area in the robotic field that studies the
behavior and how to stabilize, control, and synchronize
multiple robots for a specific task. Swarm Robotic systems
are inspired by the behavior of social animals such as
ants, bees, birds, and fish, which exemplify how simple
individual actions can generate complex social behaviors.
Swarm robotics tries to imitate these simple individual
actions to create social behaviors with evident character-
istics of robustness, scalability, and flexibility, as explained
by Camazine et al. (1999). As a result, these systems
have many potential application fields such as nano and
micro-robotics, precision agriculture, military robots, au-
tonomous army, search and rescue of victims in disaster
zones, remote sensing, mobility and transportation, and
delivery of last-mile, among others.
Over the last decades, swarm robotics have been re-
searched using different approaches to different behaviors,
such as aggregation, pattern formation, self-assembly, col-
lective exploration, coordinated motion, collective trans-
portation, consensus achievement, and task allocation.
Several research approaches around swarm robotics have
carried out experiments theoretically, using simulation en-
vironments, and in a few cases using physical robotics
platforms. This last case is the least common, given the
complexity of swarm robotics systems and the difficulty
of having physical robotic platforms. Those platforms are
needed to evaluate and verify the hypotheses and theories
of each researches field. Nevertheless, most researchers
evaluate their works through physical robotics platforms,
mostly with homogeneous terrestrial or aerial robots.
Despite the efforts to develop cheaper platforms by re-
search groups, commercial robots are highly used by re-
search projects. Commercial platforms have good advan-
tages, such as short building times or elaborated hardware
for specific tasks. However, these platforms have disadvan-
ARGroHBotS: An Affordable and
Replicable Ground Homogeneous Robot
Swarm Testbed
Nestor I. Ospina
∗
Eduardo Mojica-Nava
∗
Luis G Jaimes
∗∗
Juan M. Calder´on ∗∗∗
∗Electrical and Electronic Department, Universidad Nacional de
Colombia, Bogot´a, Colombia (e-mail: {niospinag,
eamojican}@unal.edu.co)
∗∗ Department of Computer Science, Florida Polytechnic University,
Lakeland, FL 33805, (e-mail:ljaimes@floridapoly.edu).
∗∗∗ Bethune-Cookman University, Daytona Beach, FL, USA (e-mail:
calderonj@cookman.edu), Universidad Santo Tom´as, Bogot´a, Colombia
(e-mail: juancalderon@usantotomas.edu.co)
Abstract: Currently, multi-agent mobile robotic testbeds are expensive, inaccessible, and
restricted characteristics. Besides, the remotely accessible testbeds have limitations for spe-
cific tasks and limited time of implementation. For these reasons, we designed, produced,
and implemented a testbed with excellent qualities for everybody. The ARGroHBotS is an
inexpensive, robust, accessible, and flexible testing scenario where anyone can test control and
motion theories. ARGroHBotS could be adjusted and set up for specific implementations. Due
to the ability to modify both the hardware and the software, it is possible to change both the
peripherals and the control technique used. This paper describes the design, operation, and
advantages of ARGroHBotS. This system is specifically designed as an excellent possibility to
produce an institutional platform where multi-agent and swarm algorithms could be imple-
mented and tested. The used components and technologies allow easy replication and low-cost
access, regardless of the global location. ARGroHBotS provides excellent advantages to research
groups compared with commercial Robots.
Keywords: Mechatronic Systems and Robotics, Intelligent Systems and Applications,
Sustainable Design and Control, Testbed, Multi-agent, Autonomous vehicles.
1. INTRODUCTION
In the last few years, the area of scientific research ex-
perienced dramatic growth. Moreover, the most common
way to test the advantage and qualities of new scientific
theories is by simulation software. Nevertheless, simulation
testing is enough to get a general idea of the behavior of
those theories, but no for physical implementation. For
that reason, it is necessary to use a real testing scenario
that warrants safety, the accuracy of data obtained, and
similarity to the focus environment of the research prob-
lem.
There is an area in the robotic field that studies the
behavior and how to stabilize, control, and synchronize
multiple robots for a specific task. Swarm Robotic systems
are inspired by the behavior of social animals such as
ants, bees, birds, and fish, which exemplify how simple
individual actions can generate complex social behaviors.
Swarm robotics tries to imitate these simple individual
actions to create social behaviors with evident character-
istics of robustness, scalability, and flexibility, as explained
by Camazine et al. (1999). As a result, these systems
have many potential application fields such as nano and
micro-robotics, precision agriculture, military robots, au-
tonomous army, search and rescue of victims in disaster
zones, remote sensing, mobility and transportation, and
delivery of last-mile, among others.
Over the last decades, swarm robotics have been re-
searched using different approaches to different behaviors,
such as aggregation, pattern formation, self-assembly, col-
lective exploration, coordinated motion, collective trans-
portation, consensus achievement, and task allocation.
Several research approaches around swarm robotics have
carried out experiments theoretically, using simulation en-
vironments, and in a few cases using physical robotics
platforms. This last case is the least common, given the
complexity of swarm robotics systems and the difficulty
of having physical robotic platforms. Those platforms are
needed to evaluate and verify the hypotheses and theories
of each researches field. Nevertheless, most researchers
evaluate their works through physical robotics platforms,
mostly with homogeneous terrestrial or aerial robots.
Despite the efforts to develop cheaper platforms by re-
search groups, commercial robots are highly used by re-
search projects. Commercial platforms have good advan-
tages, such as short building times or elaborated hardware
for specific tasks. However, these platforms have disadvan-
ARGroHBotS: An Affordable and
Replicable Ground Homogeneous Robot
Swarm Testbed
Nestor I. Ospina ∗Eduardo Mojica-Nava ∗Luis G Jaimes ∗∗
Juan M. Calder´on ∗∗∗
∗Electrical and Electronic Department, Universidad Nacional de
Colombia, Bogot´a, Colombia (e-mail: {niospinag,
eamojican}@unal.edu.co)
∗∗ Department of Computer Science, Florida Polytechnic University,
Lakeland, FL 33805, (e-mail:ljaimes@floridapoly.edu).
∗∗∗ Bethune-Cookman University, Daytona Beach, FL, USA (e-mail:
calderonj@cookman.edu), Universidad Santo Tom´as, Bogot´a, Colombia
(e-mail: juancalderon@usantotomas.edu.co)
Abstract: Currently, multi-agent mobile robotic testbeds are expensive, inaccessible, and
restricted characteristics. Besides, the remotely accessible testbeds have limitations for spe-
cific tasks and limited time of implementation. For these reasons, we designed, produced,
and implemented a testbed with excellent qualities for everybody. The ARGroHBotS is an
inexpensive, robust, accessible, and flexible testing scenario where anyone can test control and
motion theories. ARGroHBotS could be adjusted and set up for specific implementations. Due
to the ability to modify both the hardware and the software, it is possible to change both the
peripherals and the control technique used. This paper describes the design, operation, and
advantages of ARGroHBotS. This system is specifically designed as an excellent possibility to
produce an institutional platform where multi-agent and swarm algorithms could be imple-
mented and tested. The used components and technologies allow easy replication and low-cost
access, regardless of the global location. ARGroHBotS provides excellent advantages to research
groups compared with commercial Robots.
Keywords: Mechatronic Systems and Robotics, Intelligent Systems and Applications,
Sustainable Design and Control, Testbed, Multi-agent, Autonomous vehicles.
1. INTRODUCTION
In the last few years, the area of scientific research ex-
perienced dramatic growth. Moreover, the most common
way to test the advantage and qualities of new scientific
theories is by simulation software. Nevertheless, simulation
testing is enough to get a general idea of the behavior of
those theories, but no for physical implementation. For
that reason, it is necessary to use a real testing scenario
that warrants safety, the accuracy of data obtained, and
similarity to the focus environment of the research prob-
lem.
There is an area in the robotic field that studies the
behavior and how to stabilize, control, and synchronize
multiple robots for a specific task. Swarm Robotic systems
are inspired by the behavior of social animals such as
ants, bees, birds, and fish, which exemplify how simple
individual actions can generate complex social behaviors.
Swarm robotics tries to imitate these simple individual
actions to create social behaviors with evident character-
istics of robustness, scalability, and flexibility, as explained
by Camazine et al. (1999). As a result, these systems
have many potential application fields such as nano and
micro-robotics, precision agriculture, military robots, au-
tonomous army, search and rescue of victims in disaster
zones, remote sensing, mobility and transportation, and
delivery of last-mile, among others.
Over the last decades, swarm robotics have been re-
searched using different approaches to different behaviors,
such as aggregation, pattern formation, self-assembly, col-
lective exploration, coordinated motion, collective trans-
portation, consensus achievement, and task allocation.
Several research approaches around swarm robotics have
carried out experiments theoretically, using simulation en-
vironments, and in a few cases using physical robotics
platforms. This last case is the least common, given the
complexity of swarm robotics systems and the difficulty
of having physical robotic platforms. Those platforms are
needed to evaluate and verify the hypotheses and theories
of each researches field. Nevertheless, most researchers
evaluate their works through physical robotics platforms,
mostly with homogeneous terrestrial or aerial robots.
Despite the efforts to develop cheaper platforms by re-
search groups, commercial robots are highly used by re-
search projects. Commercial platforms have good advan-
tages, such as short building times or elaborated hardware
for specific tasks. However, these platforms have disadvan-
tages such as the high price, the difficult access depending
on the location of the research lab, and not being able
to be used for a long-range of applications or to change
spare parts (Brambilla et al., 2013). Given the need for
standard physical robotic platforms that allow generaliz-
ing the evaluation of the theories and hypotheses of the
different research areas in swarm robotics, we propose the
design and development of a complete Testbed for swarm
robotics systems with open-source hardware and software.
In other words, we propose an easily replicable robotic
platform to allow access to a more significant number of
researchers in this field of research.
2. RELATED WORK
Currently, there are several kinds of testbeds used for
training, education, and data analysis. These systems al-
low the implementation and test of a wide variety of con-
trol algorithms, coordination, consensus, route planning,
swarm stabilization, etc. Some of the earliest testbeds
are sensing testbeds, which are based on many sensors
and multiple nodes that can sense and share informa-
tion through each other. These labs have the capability
of allowing access to numerous students and researchers
remotely. Deterlab (Mirkovic and Benzel, 2012) is one of
the most popular testbeds with more than 400 general-
purpose computing nodes and a set of tools for cyberse-
curity experiments. ORBIT (Raychaudhuri et al., 2005)
allows implementing multiple algorithms remotely in a
wireless network testbed. Mobile Emulab (Johnson et al.,
2006) also uses 4 Acroname Garcia robots for mobility with
multiple wireless robots. Fit IoT-Lab (Adjih et al., 2015)
is another open access experimental IoT testbed with 2728
low-power wireless nodes and 117 mobile robots available
for numerous experiments.
Robotnacka (Petrovi and Balogh, 2012) is an educational
testbed focused on teaching and practicing a program-
ming language. It can be used for the implementation
of a considerable number of experiments. Remotely Con-
trolled Laboratory (RCL) (Gr¨ober et al., 2007) is also a
remote laboratory used to teach and experiment. Labs-
land (Ordu˜na et al., 2016) is a multiplatform laboratory
accessible through the internet that allows the implemen-
tation of multiple experiments in different labs, despite the
numerous agents. Numerous attempts have been made to
allow anyone to access a testbed remotely, for example,
PR2 (Pitzer et al., 2012) where a two-armed robot exists
with an omnidirectional base. Moreover, there is MINT
lab (De et al., 2006) where there are multiple mobile nodes
implemented using a centralized tracking system and mul-
tiple Roomba wirelessly controlled by a control server.
The University of Technology-Sidney (UTS) has been de-
veloping several remotely accessible testbeds (Kodagoda
et al., 2013). This lab only has one node that uses different
algorithms to solve the maze but cannot implement a
multi-agent experiment.
Duckietown (Paull et al., 2017) is a recent open, inex-
pensive, and flexible platform for autonomous driving,
where several autonomous driving experiments can be
implemented in multiple Duckiebots. Furthermore, this
platform can enforce a multi-lane environment, preventing
the use of algorithms that solve open or unexplored envi-
ronments. One of the famous testbeds is The Robotarium
(Wilson et al., 2020) where anyone can implement multiple
algorithms through the internet. It is necessary to simulate
the algorithm first before testing it in the Robotarium
Platform. This platform is helpful for search experiments;
however, the use is also limited due to a high number of re-
quirements. Furthermore, each robot has a fixed structure
that no allows basic modifications. Besides, the control
program of the robot cannot be modified by the public.
3. ROBOTIC PLATFORM DESIGN
The ARGroHBotS was designed, manufactured, and im-
plemented to provide swarm mobile vehicles in a safe
environment as a tool for developing different techniques
and theories in the research field. The main objective is to
provide easy, inexpensive, accessible, and open hardware,
as well as open access to each research and teaching center.
Furthermore, it was allowing customization or modifica-
tion to some components provided on this testbed. This
will let to set up an optimal environment for the task
at hand and could have differences on each differential
drive robot depending on the experiment. The main design
considerations that were taken into account are:
•Enable to be replicate independently of the worldwide
location.
•Make use of new prototyping techniques, building,
and replacing new parts at a low cost and easy to
develop.
•Integrate a robust and durable design allowing colli-
sions and hard work in multiple experiments reducing
wear and tear.
•Minimize the cost, complexity, and maintenance of
each robot.
•Provide the integration of an easy and cheap tracking
system to a testbed affordable for everyone.
•Enable easy replications and understanding of the
project providing layouts, schemes, part list, and
code.
In the research field, it is common to see how to execute a
specific experiment. Still, the integrity of the implementa-
tion depends on how close the test is to the physical work.
For this reason, it is necessary to use an open hardware
platform where each laboratory could modify, adding or
improving it to focus on different tasks. In addition, it is
vital to recognize that there is a necessity for an open-
access laboratory where it could always be available to
implement and obtaining feedback in real-time on each
research project remotely.
3.1 Mobile Robot
The robot design had to account for the different mod-
els made until now. Some of them are Kilobot (Ruben-
stein et al., 2012), Jasmine (Kernbach et al., 2009), Alice
(Caprari et al., 1998), R-one (McLurkin et al., 2006),
SwarmBot (McLurkin et al., 2013), E-puck (Mondada
et al., 2009), AERobot (Rubenstein et al., 2015), LabRat
(Robinette et al., 2009), Thymio II (Riedo et al., 2013),
DuckieBot (Paull et al., 2017), Create 2 (Dekan et al.,
2013), and GRITSBot (Pickem et al., 2015).
Considering the previous robots and their design condi-
tions, the BeCBot was made with multiple accessible and

Nestor I. Ospina et al. / IFAC PapersOnLine 54-13 (2021) 256–261 257
Copyright ©
2021 The Authors. This is an open access article under the CC BY-NC-ND license
(
https://creativecommons.org/licenses/by-nc-nd/4.0/
)
tages such as the high price, the difficult access depending
on the location of the research lab, and not being able
to be used for a long-range of applications or to change
spare parts (Brambilla et al., 2013). Given the need for
standard physical robotic platforms that allow generaliz-
ing the evaluation of the theories and hypotheses of the
different research areas in swarm robotics, we propose the
design and development of a complete Testbed for swarm
robotics systems with open-source hardware and software.
In other words, we propose an easily replicable robotic
platform to allow access to a more significant number of
researchers in this field of research.
2. RELATED WORK
Currently, there are several kinds of testbeds used for
training, education, and data analysis. These systems al-
low the implementation and test of a wide variety of con-
trol algorithms, coordination, consensus, route planning,
swarm stabilization, etc. Some of the earliest testbeds
are sensing testbeds, which are based on many sensors
and multiple nodes that can sense and share informa-
tion through each other. These labs have the capability
of allowing access to numerous students and researchers
remotely. Deterlab (Mirkovic and Benzel, 2012) is one of
the most popular testbeds with more than 400 general-
purpose computing nodes and a set of tools for cyberse-
curity experiments. ORBIT (Raychaudhuri et al., 2005)
allows implementing multiple algorithms remotely in a
wireless network testbed. Mobile Emulab (Johnson et al.,
2006) also uses 4 Acroname Garcia robots for mobility with
multiple wireless robots. Fit IoT-Lab (Adjih et al., 2015)
is another open access experimental IoT testbed with 2728
low-power wireless nodes and 117 mobile robots available
for numerous experiments.
Robotnacka (Petrovi and Balogh, 2012) is an educational
testbed focused on teaching and practicing a program-
ming language. It can be used for the implementation
of a considerable number of experiments. Remotely Con-
trolled Laboratory (RCL) (Gr¨ober et al., 2007) is also a
remote laboratory used to teach and experiment. Labs-
land (Ordu˜na et al., 2016) is a multiplatform laboratory
accessible through the internet that allows the implemen-
tation of multiple experiments in different labs, despite the
numerous agents. Numerous attempts have been made to
allow anyone to access a testbed remotely, for example,
PR2 (Pitzer et al., 2012) where a two-armed robot exists
with an omnidirectional base. Moreover, there is MINT
lab (De et al., 2006) where there are multiple mobile nodes
implemented using a centralized tracking system and mul-
tiple Roomba wirelessly controlled by a control server.
The University of Technology-Sidney (UTS) has been de-
veloping several remotely accessible testbeds (Kodagoda
et al., 2013). This lab only has one node that uses different
algorithms to solve the maze but cannot implement a
multi-agent experiment.
Duckietown (Paull et al., 2017) is a recent open, inex-
pensive, and flexible platform for autonomous driving,
where several autonomous driving experiments can be
implemented in multiple Duckiebots. Furthermore, this
platform can enforce a multi-lane environment, preventing
the use of algorithms that solve open or unexplored envi-
ronments. One of the famous testbeds is The Robotarium
(Wilson et al., 2020) where anyone can implement multiple
algorithms through the internet. It is necessary to simulate
the algorithm first before testing it in the Robotarium
Platform. This platform is helpful for search experiments;
however, the use is also limited due to a high number of re-
quirements. Furthermore, each robot has a fixed structure
that no allows basic modifications. Besides, the control
program of the robot cannot be modified by the public.
3. ROBOTIC PLATFORM DESIGN
The ARGroHBotS was designed, manufactured, and im-
plemented to provide swarm mobile vehicles in a safe
environment as a tool for developing different techniques
and theories in the research field. The main objective is to
provide easy, inexpensive, accessible, and open hardware,
as well as open access to each research and teaching center.
Furthermore, it was allowing customization or modifica-
tion to some components provided on this testbed. This
will let to set up an optimal environment for the task
at hand and could have differences on each differential
drive robot depending on the experiment. The main design
considerations that were taken into account are:
•Enable to be replicate independently of the worldwide
location.
•Make use of new prototyping techniques, building,
and replacing new parts at a low cost and easy to
develop.
•Integrate a robust and durable design allowing colli-
sions and hard work in multiple experiments reducing
wear and tear.
•Minimize the cost, complexity, and maintenance of
each robot.
•Provide the integration of an easy and cheap tracking
system to a testbed affordable for everyone.
•Enable easy replications and understanding of the
project providing layouts, schemes, part list, and
code.
In the research field, it is common to see how to execute a
specific experiment. Still, the integrity of the implementa-
tion depends on how close the test is to the physical work.
For this reason, it is necessary to use an open hardware
platform where each laboratory could modify, adding or
improving it to focus on different tasks. In addition, it is
vital to recognize that there is a necessity for an open-
access laboratory where it could always be available to
implement and obtaining feedback in real-time on each
research project remotely.
3.1 Mobile Robot
The robot design had to account for the different mod-
els made until now. Some of them are Kilobot (Ruben-
stein et al., 2012), Jasmine (Kernbach et al., 2009), Alice
(Caprari et al., 1998), R-one (McLurkin et al., 2006),
SwarmBot (McLurkin et al., 2013), E-puck (Mondada
et al., 2009), AERobot (Rubenstein et al., 2015), LabRat
(Robinette et al., 2009), Thymio II (Riedo et al., 2013),
DuckieBot (Paull et al., 2017), Create 2 (Dekan et al.,
2013), and GRITSBot (Pickem et al., 2015).
Considering the previous robots and their design condi-
tions, the BeCBot was made with multiple accessible and

258 Nestor I. Ospina et al. / IFAC PapersOnLine 54-13 (2021) 256–261
inexpensive parts that make this robot robust and efficient.
Additionally, it can be replicated at a low cost due to the
easiness of 3D printed pieces that composed the majority
of the robot. Fig. 1 shows our robot design. It is designed
based on a commercial board. The frame and cover were
designed to be printed in a 3D machine.
Fig. 1. Exploded Robot
One of the advantages of this design is that it could be
downloaded ((Ospina and Calderon, 2020)) and be mod-
ified simply by changing the CAD. In Fig 2 it is possible
to observe the CAD robot model and the picture of a real
robot.
Fig. 2. Differential Drive Robot.
kinematic model The Kinematic model of the BeCBot is
a unicycle type (Carona et al., 2008). A simple non-linear
model usually describes this model
˙x=vcos θ
˙y=vsin θ
˙
θ=ω
(1)
where x, y, θ are the position and orientation in the world
reference, and v, w is the linear and angular velocity,
respectively. The last are the control variables of the robot.
Fig 3 shows the representation of the mentioned model.
The Robot Processing system receives the control variables
v, w and, taking into account the radius and the distance
between its wheels, compute the angular velocity of each
wheel wr,w
l, using the equation 2.
Fr
Fl=1
2R2−L
2Lv
w.(2)
The microprocessor has an internal closed-loop control
that senses the real position of each robot wheel and
Fig. 3. Kinematic model
controls both the position and velocity to ensure the
desired velocity. The value of distance between wheels is
L= 12 [cm], the radius of each wheel isR= 3 [cm], and
the mass m= 709 [gr]. These constant values are fixed in
the control law.
Robot Processing System The Robot processing system
was equipped with an Auriga board. It contains a micro-
controller Atmega 2560 running at a 16 MHZ speed clock.
In addition, the developing board has Gyroscope, ther-
mistor, and a sound sensor for perception applications. It
also has a Bluetooth shield that allows communication and
reprogramming the microprocessor, but it can be replaced
for RF or Wifi modules. Moreover, it has 9 I2C ports, 1
UART port, and one port for smart motor applications.
Besides, the mainboard controls the encoder motors with
a TB6612 chip. With the integration of this chip, the
microprocessor can control with high precision the velocity
and position of each engine with an internal control closed
loop.
Communication System The communication system
considers the number of nodes and the bandwidth of the
shared information. As a result, the ESP 8266 module
was chosen due to its low-cost price, affordable and ease
of programming. The improvement here is the network
protocol named ESP-NOW. It can connect, transmit, and
receive up to a 250-byte payload by WiFi protocol. This
new protocol supports unencrypted peers; however, their
total number should be less than 20, including encrypted
peers. This protocol allows sharing information without
the necessity of a WiFi station or router as a principal
module of interconnection. Besides, with this protocol, if
suddenly one of the boards loses power or resets, it will
automatically connect to its peer to continue the commu-
nication when it restarts.
3.2 Global Position system
The ARGroHBotS is equipped with a fiducial marker
system for camera pose estimation (Garrido-Jurado et al.,
2014). This system uses a web camera and system image
processing to detect and estimate the position of each
robot. This is currently implemented using the OpenCV
library in python language in the ArUco markers. The
system can calculate the x,y,z, and θposition of each
agent. The platform uses a Logitech C930 at a resolution
of 1920x1080 pixels and a 90-degree field-of-view. It can
achieve frame rates of up to 30 fps. This library allows the
detection of up to 1000 tags in the same frame.
Power system The BeCBot has a lithium-ion battery of
3000 mAh. It provides energy for about ten working hours.
Moreover, the charging system recharges the battery in
about 2 hours. In addition, this battery can provide up to
3 amps in a barrel and USB ports. This makes the power
system robust and valuable to implement with different
development boards.
3.3 The Interconnect System
The ARGroHBotS was built with the integration of multi-
ple communication, tracking, processing, and sensing sys-
tems. All of them can be replaced in the future for better
technologies or another system that works better.
The tracking system uses the OpenCV library and ArUco
tags. The camera takes multiple frames to process the
information, estimating the positions x,y, and orientation.
It gives a vector p∈R2,nr, where nris the number
of robots in each experiment. With this information,
the central processing unit computes the value of the
robot’s control variables using the preferred technique.
Then it broadcast this information to each mobile robot
via WiFi protocol and if the test needs it, waits for
feedback information. Eventually, each robot gets data
from the server, and each one takes a decision. The server
can record, graph, and save all the information that was
taken in each test. Additionally, the ARGroHBotS could
implement reactive, nonreactive, path following, and other
navigation techniques. In Figure 4 is shown the connection
of the entire testbed and how each section interacts.
Fig. 4. System Architecture
4. EXPERIMENTAL TEST AND SYSTEM
EVALUATION
Currently, the ARGroHBotS has been tested with three
different algorithms. It was obtained the expected results
thanks to the flexible platform it can be used for various
applications. The theories used are swarm navigation,
Event-Triggered Control, and consensus. Figure 5 shows
the environment where the experiments were carried out.
Fig. 5. Picture of The ARGroHBotS implementing an
experiment with 4 BeCBots
4.1 Robot Swarm Navigation
This algorithm was implemented using stochastic naviga-
tion theory. This is taken from the papers Cardona and
Calderon (2019) and Le´on et al. (2016). The control law
is based on a repulsion function.
−kaxij (3)
The repulsion parameter kaallows the agent to keep a
comfortable distance xij from the other swarm members
and avoid collision between agents when the swarm is mov-
ing. The repulsion mechanism can react in two different
ways. Initially, when a comfortable distance is given, and
its parameter is revealed in the following equation:
[−k(xij −d)](xij ).(4)
Where xij =2
(xi−xj)T(xi−xj), k>0 is the
repulsion magnitude and dis the comfortable distance
between ith agent and the jth agent. On the other hand,
if two agents are two close to each other, the parameter is
represented in the next equation.
krexp(−1
2xij 2
r2
s
)(xij ),(5)
where kr>0, represents the repulsion magnitude and
rs>0, is the repulsion range. The figure 6 shows the
trajectory of each robot navigation interacting with its
neighbors.
Fig. 6. Robot swarm navigation experiment

Nestor I. Ospina et al. / IFAC PapersOnLine 54-13 (2021) 256–261 259
achieve frame rates of up to 30 fps. This library allows the
detection of up to 1000 tags in the same frame.
Power system The BeCBot has a lithium-ion battery of
3000 mAh. It provides energy for about ten working hours.
Moreover, the charging system recharges the battery in
about 2 hours. In addition, this battery can provide up to
3 amps in a barrel and USB ports. This makes the power
system robust and valuable to implement with different
development boards.
3.3 The Interconnect System
The ARGroHBotS was built with the integration of multi-
ple communication, tracking, processing, and sensing sys-
tems. All of them can be replaced in the future for better
technologies or another system that works better.
The tracking system uses the OpenCV library and ArUco
tags. The camera takes multiple frames to process the
information, estimating the positions x,y, and orientation.
It gives a vector p∈R2,nr, where nris the number
of robots in each experiment. With this information,
the central processing unit computes the value of the
robot’s control variables using the preferred technique.
Then it broadcast this information to each mobile robot
via WiFi protocol and if the test needs it, waits for
feedback information. Eventually, each robot gets data
from the server, and each one takes a decision. The server
can record, graph, and save all the information that was
taken in each test. Additionally, the ARGroHBotS could
implement reactive, nonreactive, path following, and other
navigation techniques. In Figure 4 is shown the connection
of the entire testbed and how each section interacts.
Fig. 4. System Architecture
4. EXPERIMENTAL TEST AND SYSTEM
EVALUATION
Currently, the ARGroHBotS has been tested with three
different algorithms. It was obtained the expected results
thanks to the flexible platform it can be used for various
applications. The theories used are swarm navigation,
Event-Triggered Control, and consensus. Figure 5 shows
the environment where the experiments were carried out.
Fig. 5. Picture of The ARGroHBotS implementing an
experiment with 4 BeCBots
4.1 Robot Swarm Navigation
This algorithm was implemented using stochastic naviga-
tion theory. This is taken from the papers Cardona and
Calderon (2019) and Le´on et al. (2016). The control law
is based on a repulsion function.
−kaxij (3)
The repulsion parameter kaallows the agent to keep a
comfortable distance xij from the other swarm members
and avoid collision between agents when the swarm is mov-
ing. The repulsion mechanism can react in two different
ways. Initially, when a comfortable distance is given, and
its parameter is revealed in the following equation:
[−k(xij −d)](xij ).(4)
Where xij =2
(xi−xj)T(xi−xj), k>0 is the
repulsion magnitude and dis the comfortable distance
between ith agent and the jth agent. On the other hand,
if two agents are two close to each other, the parameter is
represented in the next equation.
krexp(−1
2xij 2
r2
s
)(xij ),(5)
where kr>0, represents the repulsion magnitude and
rs>0, is the repulsion range. The figure 6 shows the
trajectory of each robot navigation interacting with its
neighbors.
Fig. 6. Robot swarm navigation experiment

260 Nestor I. Ospina et al. / IFAC PapersOnLine 54-13 (2021) 256–261
4.2 Event-Triggered Control
An Event-triggered Control was implemented based on
Event-triggered coordination of multi-agent systems in
agricultural environments (Pabon and Mojica-Nava, 2019).
This is based on a leader-follower topology, which allows
that the leader robot knows the path; the other agents have
limited and selected connections with the leading agent
and with each other (Pabon et al., 2021). The following
equation can represent this behavior,
˙xi=
j∈N (i)
(xj(t)−xi(t)),(6)
where Nis the set of robots that are close to each robot.
xiis the global position of each BeCBot, and xjis the
position of its neighbors. Using the mentioned concepts,
the entire behavior of the system can be represented as
˙x(t)=−L(G)x(t).
The results are shown by Figure 7. The fourth BeCBot
has an error with the reference due to the topology and
the lost communication, but it is an expected result
Fig. 7. ETC implementation, with path traveled and
reference used.
Video: https://youtu.be/mvQeEzIGbgQ
4.3 Consensus
Another algorithm implemented in the lab was the multi-
agent consensus using the control theory of the paper
Mesbahi and Egerstedt (2010). In this, each agent uses a
control law to converge, using the following control input
˙xi=
j∈N (i)
(xj(t)−xi(t)) + d, (7)
where dis the distance between robots, this to prevent
collisions and allow the formation consensus as depicted
by Cardona et al. (2021). Figure 8 shows how the BeCBots
start the navigation in a random position and taking
interaction with its neighbors achieve to make a forma-
tion around the center of the testbed. Implementing this
reactive navigation, we obtained the expected results
5. CONCLUSION
In this paper, we described the development of an open-
source multi-robot platform. It aims to be shared with
the research and teaching areas where it can be useful.
So far, robot swarm navigation, consensus, and event-
triggered control algorithms have been implemented in this
Fig. 8. The consensus experiment
platform. Ood results were obtained, which prove that the
platform is versatile. Results allow verifying, evaluating,
and implementing several algorithms used in mobile multi-
agent systems.
This testbed was designed and developed with a differ-
ential mobile robot, but we plan to extend it to different
mobile architectures to work together in the same environ-
ment. It will allow improvements in heterogeneous swarm
research.
REFERENCES
Adjih, C., Baccelli, E., Fleury, E., Harter, G., Mitton,
N., Noel, T., Pissard-Gibollet, R., Saint-Marcel, F.,
Schreiner, G., Vandaele, J., et al. (2015). Fit iot-lab:
A large scale open experimental iot testbed. In 2015
IEEE 2nd World Forum on Internet of Things (WF-
IoT), 459–464. IEEE.
Brambilla, M., Ferrante, E., Birattari, M., and Dorigo,
M. (2013). Swarm robotics: a review from the swarm
engineering perspective. Swarm Intelligence, 7(1), 1–41.
Camazine, S., Visscher, P.K., Finley, J., and Vetter, R.S.
(1999). House-hunting by honey bee swarms: collective
decisions and individual behaviors. Insectes Sociaux,
46(4), 348–360.
Caprari, G., Balmer, P., Piguet, R., and Siegwart, R.
(1998). The autonomous micro robot” alice”: a plat-
form for scientific and commercial applications. In
MHA’98. Proceedings of the 1998 International Sym-
posium on Micromechatronics and Human Science.-
Creation of New Industry-(Cat. No. 98TH8388), 231–
235. IEEE.
Cardona, G.A. and Calderon, J.M. (2019). Robot swarm
navigation and victim detection using rendezvous con-
sensus in search and rescue operations. Applied Sciences,
9(8), 1702.
Cardona, G.A., Ramirez-Rugeles, J., Mojica-Nava, E., and
Calderon, J.M. (2021). Visual victim detection and
quadrotor-swarm coordination control in search and
rescue environment. International Journal of Electrical
& Computer Engineering (2088-8708), 11(3).
Carona, R., Aguiar, A.P., and Gaspar, J. (2008). Control
of unicycle type robots tracking, path following and
point stabilization.
De, P., Raniwala, A., Krishnan, R., Tatavarthi, K., Modi,
J., Syed, N.A., Sharma, S., and Chiueh, T.c. (2006).
Mint-m: an autonomous mobile wireless experimenta-
tion platform. In Proceedings of the 4th international
conference on Mobile systems, applications and services,
124–137.
Dekan, M., Duchon, F., Jurisica, L., Vitko, A., and
Babinec, A. (2013). irobot create used in education.
Journal of Mechanics Engineering and Automation,
3(4), 197–202.
Garrido-Jurado, S., Mu˜noz-Salinas, R., Madrid-Cuevas,
F.J., and Mar´ın-Jim´enez, M.J. (2014). Automatic gen-
eration and detection of highly reliable fiducial markers
under occlusion. Pattern Recognition, 47(6), 2280–2292.
Gr¨ober, S., Vetter, M., Eckert, B., and Jodl, H.J. (2007).
Experimenting from a distance—remotely controlled
laboratory (rcl). European Journal of Physics, 28(3),
S127.
Johnson, D., Stack, T., Fish, R., Flickinger, D.M., Stoller,
L., Ricci, R., and Lepreau, J. (2006). Mobile emulab: A
robotic wireless and sensor network testbed. In Proceed-
ings IEEE INFOCOM 2006. 25TH IEEE International
Conference on Computer Communications, 1–12. IEEE.
Kernbach, S., Thenius, R., Kernbach, O., and Schmickl,
T. (2009). Re-embodiment of honeybee aggregation
behavior in an artificial micro-robotic system. Adaptive
Behavior, 17(3), 237–259.
Kodagoda, S., Alempijevic, A., Huang, S., de la Villefro-
moy, M., Diponio, M., and Cogar, L. (2013). Mov-
ing away from simulations: Innovative assessment of
mechatronic subjects using remote laboratories. In 2013
12th International Conference on Information Technol-
ogy Based Higher Education and Training (ITHET), 1–
5. IEEE.
Le´on, J., Cardona, G.A., Botello, A., and Calder´on, J.M.
(2016). Robot swarms theory applicable to seek and
rescue operation. In International Conference on In-
telligent Systems Design and Applications, 1061–1070.
Springer.
McLurkin, J., Lynch, A.J., Rixner, S., Barr, T.W., Chou,
A., Foster, K., and Bilstein, S. (2013). A low-cost
multi-robot system for research, teaching, and outreach.
In Distributed Autonomous Robotic Systems, 597–609.
Springer.
McLurkin, J., Smith, J., Frankel, J., Sotkowitz, D., Blau,
D., and Schmidt, B. (2006). Speaking swarmish:
Human-robot interface design for large swarms of au-
tonomous mobile robots. In AAAI spring symposium:
to boldly go where no human-robot team has gone before,
volume 72. Palo Alto, CA.
Mesbahi, M. and Egerstedt, M. (2010). Graph theoretic
methods in multiagent networks, volume 33. Princeton
University Press.
Mirkovic, J. and Benzel, T. (2012). Teaching cybersecurity
with deterlab. IEEE Security Privacy, 10(01), 73–76.
doi:10.1109/MSP.2012.23.
Mondada, F., Bonani, M., Raemy, X., Pugh, J., Cianci,
C., Klaptocz, A., Magnenat, S., Zufferey, J.C., Floreano,
D., and Martinoli, A. (2009). The e-puck, a robot
designed for education in engineering. In Proceedings
of the 9th conference on autonomous robot systems
and competitions, volume 1, 59–65. IPCB: Instituto
Polit´ecnico de Castelo Branco.
Ordu˜na, P., Rodriguez-Gil, L., Garcia-Zubia, J., Angulo,
I., Hernandez, U., and Azcuenaga, E. (2016). Labsland:
A sharing economy platform to promote educational
remote laboratories maintainability, sustainability and
adoption. In 2016 IEEE Frontiers in Education Confer-
ence (FIE), 1–6. doi:10.1109/FIE.2016.7757579.
Ospina, N.I. and Calderon, J. (2020). Becbot. url-
https://www.thingiverse.com/thing:4646445.
Pabon, J.D. and Mojica-Nava, E. (2019). Event-triggered
coordination of multi-agent systems in agricultural envi-
ronments. In 2019 IEEE 4th Colombian Conference on
Automatic Control (CCAC), 1–6. doi:10.1109/CCAC.
2019.8920970.
Pabon, J.D., Cardona, G.A., Ospina, N.I., Calderon, J.M.,
and Mojica-Nava, E. (2021). Event-triggered control
for weight-unbalanced directed robot networks. In
2021 IEEE/RSJ International Conference on Intelligent
Robots and Systems (IROS). IEEE.
Paull, L., Tani, J., Ahn, H., Alonso-Mora, J., Carlone, L.,
Cap, M., Chen, Y.F., Choi, C., Dusek, J., Fang, Y.,
et al. (2017). Duckietown: an open, inexpensive and
flexible platform for autonomy education and research.
In 2017 IEEE International Conference on Robotics and
Automation (ICRA), 1497–1504. IEEE.
Petrovi, P. and Balogh, R. (2012). Deployment of
remotely-accessible robotics laboratory. International
Journal of Online and Biomedical Engineering (iJOE),
8(S2), 31–35.
Pickem, D., Lee, M., and Egerstedt, M. (2015). The
gritsbot in its natural habitat-a multi-robot testbed. In
2015 IEEE International Conference on Robotics and
Automation (ICRA), 4062–4067. IEEE.
Pitzer, B., Osentoski, S., Jay, G., Crick, C., and Jenkins,
O.C. (2012). Pr2 remote lab: An environment for re-
mote development and experimentation. In 2012 IEEE
International Conference on Robotics and Automation,
3200–3205. IEEE.
Raychaudhuri, D., Seskar, I., Ott, M., Ganu, S., Ra-
machandran, K., Kremo, H., Siracusa, R., Liu, H., and
Singh, M. (2005). Overview of the orbit radio grid
testbed for evaluation of next-generation wireless net-
work protocols. In IEEE Wireless Communications
and Networking Conference, 2005, volume 3, 1664–1669.
IEEE.
Riedo, F., Chevalier, M., Magnenat, S., and Mondada,
F. (2013). Thymio ii, a robot that grows wiser with
children. In 2013 IEEE Workshop on Advanced Robotics
and its Social Impacts, 187–193. IEEE.
Robinette, P., Meuth, R., Dolan, R., and Wunsch, D.
(2009). Labrat™: Miniature robot for students, re-
searchers, and hobbyists. In 2009 IEEE/RSJ Inter-
national Conference on Intelligent Robots and Systems,
1007–1012. IEEE.
Rubenstein, M., Ahler, C., and Nagpal, R. (2012). Kilobot:
A low cost scalable robot system for collective behaviors.
In 2012 IEEE International Conference on Robotics
and Automation, 3293–3298. doi:10.1109/ICRA.2012.
6224638.
Rubenstein, M., Cimino, B., Nagpal, R., and Werfel, J.
(2015). Aerobot: An affordable one-robot-per-student
system for early robotics education. In 2015 IEEE
International Conference on Robotics and Automation
(ICRA), 6107–6113. IEEE.
Wilson, S., Glotfelter, P., Wang, L., Mayya, S., Notomista,
G., Mote, M., and Egerstedt, M. (2020). The robotar-
ium: Globally impactful opportunities, challenges, and
lessons learned in remote-access, distributed control of
multirobot systems. IEEE Control Systems Magazine,
40(1), 26–44. doi:10.1109/MCS.2019.2949973.

Nestor I. Ospina et al. / IFAC PapersOnLine 54-13 (2021) 256–261 261
Dekan, M., Duchon, F., Jurisica, L., Vitko, A., and
Babinec, A. (2013). irobot create used in education.
Journal of Mechanics Engineering and Automation,
3(4), 197–202.
Garrido-Jurado, S., Mu˜noz-Salinas, R., Madrid-Cuevas,
F.J., and Mar´ın-Jim´enez, M.J. (2014). Automatic gen-
eration and detection of highly reliable fiducial markers
under occlusion. Pattern Recognition, 47(6), 2280–2292.
Gr¨ober, S., Vetter, M., Eckert, B., and Jodl, H.J. (2007).
Experimenting from a distance—remotely controlled
laboratory (rcl). European Journal of Physics, 28(3),
S127.
Johnson, D., Stack, T., Fish, R., Flickinger, D.M., Stoller,
L., Ricci, R., and Lepreau, J. (2006). Mobile emulab: A
robotic wireless and sensor network testbed. In Proceed-
ings IEEE INFOCOM 2006. 25TH IEEE International
Conference on Computer Communications, 1–12. IEEE.
Kernbach, S., Thenius, R., Kernbach, O., and Schmickl,
T. (2009). Re-embodiment of honeybee aggregation
behavior in an artificial micro-robotic system. Adaptive
Behavior, 17(3), 237–259.
Kodagoda, S., Alempijevic, A., Huang, S., de la Villefro-
moy, M., Diponio, M., and Cogar, L. (2013). Mov-
ing away from simulations: Innovative assessment of
mechatronic subjects using remote laboratories. In 2013
12th International Conference on Information Technol-
ogy Based Higher Education and Training (ITHET), 1–
5. IEEE.
Le´on, J., Cardona, G.A., Botello, A., and Calder´on, J.M.
(2016). Robot swarms theory applicable to seek and
rescue operation. In International Conference on In-
telligent Systems Design and Applications, 1061–1070.
Springer.
McLurkin, J., Lynch, A.J., Rixner, S., Barr, T.W., Chou,
A., Foster, K., and Bilstein, S. (2013). A low-cost
multi-robot system for research, teaching, and outreach.
In Distributed Autonomous Robotic Systems, 597–609.
Springer.
McLurkin, J., Smith, J., Frankel, J., Sotkowitz, D., Blau,
D., and Schmidt, B. (2006). Speaking swarmish:
Human-robot interface design for large swarms of au-
tonomous mobile robots. In AAAI spring symposium:
to boldly go where no human-robot team has gone before,
volume 72. Palo Alto, CA.
Mesbahi, M. and Egerstedt, M. (2010). Graph theoretic
methods in multiagent networks, volume 33. Princeton
University Press.
Mirkovic, J. and Benzel, T. (2012). Teaching cybersecurity
with deterlab. IEEE Security Privacy, 10(01), 73–76.
doi:10.1109/MSP.2012.23.
Mondada, F., Bonani, M., Raemy, X., Pugh, J., Cianci,
C., Klaptocz, A., Magnenat, S., Zufferey, J.C., Floreano,
D., and Martinoli, A. (2009). The e-puck, a robot
designed for education in engineering. In Proceedings
of the 9th conference on autonomous robot systems
and competitions, volume 1, 59–65. IPCB: Instituto
Polit´ecnico de Castelo Branco.
Ordu˜na, P., Rodriguez-Gil, L., Garcia-Zubia, J., Angulo,
I., Hernandez, U., and Azcuenaga, E. (2016). Labsland:
A sharing economy platform to promote educational
remote laboratories maintainability, sustainability and
adoption. In 2016 IEEE Frontiers in Education Confer-
ence (FIE), 1–6. doi:10.1109/FIE.2016.7757579.
Ospina, N.I. and Calderon, J. (2020). Becbot. url-
https://www.thingiverse.com/thing:4646445.
Pabon, J.D. and Mojica-Nava, E. (2019). Event-triggered
coordination of multi-agent systems in agricultural envi-
ronments. In 2019 IEEE 4th Colombian Conference on
Automatic Control (CCAC), 1–6. doi:10.1109/CCAC.
2019.8920970.
Pabon, J.D., Cardona, G.A., Ospina, N.I., Calderon, J.M.,
and Mojica-Nava, E. (2021). Event-triggered control
for weight-unbalanced directed robot networks. In
2021 IEEE/RSJ International Conference on Intelligent
Robots and Systems (IROS). IEEE.
Paull, L., Tani, J., Ahn, H., Alonso-Mora, J., Carlone, L.,
Cap, M., Chen, Y.F., Choi, C., Dusek, J., Fang, Y.,
et al. (2017). Duckietown: an open, inexpensive and
flexible platform for autonomy education and research.
In 2017 IEEE International Conference on Robotics and
Automation (ICRA), 1497–1504. IEEE.
Petrovi, P. and Balogh, R. (2012). Deployment of
remotely-accessible robotics laboratory. International
Journal of Online and Biomedical Engineering (iJOE),
8(S2), 31–35.
Pickem, D., Lee, M., and Egerstedt, M. (2015). The
gritsbot in its natural habitat-a multi-robot testbed. In
2015 IEEE International Conference on Robotics and
Automation (ICRA), 4062–4067. IEEE.
Pitzer, B., Osentoski, S., Jay, G., Crick, C., and Jenkins,
O.C. (2012). Pr2 remote lab: An environment for re-
mote development and experimentation. In 2012 IEEE
International Conference on Robotics and Automation,
3200–3205. IEEE.
Raychaudhuri, D., Seskar, I., Ott, M., Ganu, S., Ra-
machandran, K., Kremo, H., Siracusa, R., Liu, H., and
Singh, M. (2005). Overview of the orbit radio grid
testbed for evaluation of next-generation wireless net-
work protocols. In IEEE Wireless Communications
and Networking Conference, 2005, volume 3, 1664–1669.
IEEE.
Riedo, F., Chevalier, M., Magnenat, S., and Mondada,
F. (2013). Thymio ii, a robot that grows wiser with
children. In 2013 IEEE Workshop on Advanced Robotics
and its Social Impacts, 187–193. IEEE.
Robinette, P., Meuth, R., Dolan, R., and Wunsch, D.
(2009). Labrat™: Miniature robot for students, re-
searchers, and hobbyists. In 2009 IEEE/RSJ Inter-
national Conference on Intelligent Robots and Systems,
1007–1012. IEEE.
Rubenstein, M., Ahler, C., and Nagpal, R. (2012). Kilobot:
A low cost scalable robot system for collective behaviors.
In 2012 IEEE International Conference on Robotics
and Automation, 3293–3298. doi:10.1109/ICRA.2012.
6224638.
Rubenstein, M., Cimino, B., Nagpal, R., and Werfel, J.
(2015). Aerobot: An affordable one-robot-per-student
system for early robotics education. In 2015 IEEE
International Conference on Robotics and Automation
(ICRA), 6107–6113. IEEE.
Wilson, S., Glotfelter, P., Wang, L., Mayya, S., Notomista,
G., Mote, M., and Egerstedt, M. (2020). The robotar-
ium: Globally impactful opportunities, challenges, and
lessons learned in remote-access, distributed control of
multirobot systems. IEEE Control Systems Magazine,
40(1), 26–44. doi:10.1109/MCS.2019.2949973.