Short Paper—Remote Lab for Robotics Applications
Remote Lab for Robotics Applications
[Link]
Robinson Jiménez M.!!", Oscar F. Avilés S, Mauricio F. Mauledoux M.
Nueva Granada Military University, Bogotá, Colombia
[Link]@[Link]
Abstract—This article describes the development of a remote lab environ-
ment used to test and training sessions for robotics tasks. This environment is
made up of the components and devices based on two robotic arms, a network
link, Arduino card and Arduino shield for Ethernet, as well as an IP camera.
The remote laboratory is implemented to perform remote control of the robotic
arms with visual feedback by camera, of the robots actions, where, with a group
of test users, it was possible to obtain performance ranges in tasks of telecontrol
of up to 92%.
Keywords—remote environment, remote lab, robotic arm, robotics training
1 Introduction
With the increasing boom of robotics in all human activities, being familiar with
the control of a robotic agent is not unusual. Robotic toys today are very common, it
is possible to find from assistive robots to support feeding to people with limitations
[1] to assistant robots of physical activity for adolescents [2], which are part of the
great variety of developments found in the state of the art. The presentations of the
robotic agents are very varied, until reaching critical support points, as intermediaries
in teleoperations [3]. This type of applications allows to see the incidence of the ro-
bots in the activities of the man and his improvement of the quality of life.
Due to this, to develop platforms of training in the control of robots, versatile for
any age range, becomes important for its assertive use and development. Nowadays,
with the possibility of remote control of different systems, such as air conditioning
[4], control of robots [5] and similar, coupled with the advantages of virtual laborato-
ries [6], remote laboratories for telecontrol robots offer solutions for needs such as
training in various areas [7], for example in the case of teleoperation.
Remote laboratory environments already carry several developments reported in
the state of the art [8][9], and have shown several advantages, such as low cost im-
plementation [10][11] and skill development in robot control and programming algo-
rithms [12][13]. In this article, it is presented a remote laboratory environment for
telecontrol of two robotic arm type agents, capable of interacting with each other,
using visual feedback from the user through an IP camera, looking for a simple envi-
ronment of easy assimilation, for learning and training in robot telecontrol.
iJOE ‒ Vol. 14, No. 1, 2018 187
Short Paper—Remote Lab for Robotics Applications
Then the article is divided into 4 sections, section two describes the design of the
remote laboratory, section three presents the analysis of results and finally section
four gives the conclusions reached.
2 Application environment
The proposed remote laboratory consists of the manipulation of two robotic arms
to tele-command, a monitoring camera for visual feedback and an Arduino card with
expander card for Ethernet connectivity, network access routers are used to connect
the remote site with the laboratory. In the remote site there is a web interface for ma-
nipulation of each degree of freedom by robotic agent and setting of the connection.
This scheme, which proposes a bidirectional communication, is shown in figure 1.
Fig. 1. Application environment.
The first aspect required is to do the kinematic analysis of the robot with respect to
a reference system, which allows to determine the position and orientation of the end
effector of the robot, known the values of the joints and the geometric parameters of
the elements of the robot. For this analysis the Denavit-Hartenberg (D-H) algorithm is
used, which is based on the assignment of axes coordinated by degrees of freedom as
shown in Figure 1, both robotic arms used have the same architecture.
The kinematic analysis, as shown in figure 2, is derived to determine the D-H ma-
trix, which allows to relate the spatial variations of each degree of freedom and to
concatenate them. The virtual environment will allow manipulating each degree of
freedom, for the rotational case, using the D-H matrix shown in Table 1.
188 [Link]
Short Paper—Remote Lab for Robotics Applications
Fig. 2. Inverse Kinematics.
Table 1. Kinematic matrix
Joint ! d a "
1 q1 d1 0 -90
2 q2 0 d2 0
3 q3 0 d3 0
4 q4-90 0 d4 -90
5 q5 d5 0 0
The energization of all degree of freedom, controlled by servomotor, requires a
regulated source of 12 volts at a current of 5 amperes, for the adequate manipulation
of both arms, in total 12 servomotors, the power supply provides 12 volts and a max-
imum current of 5 amps. A voltage regulator L78S05 is used for supplying the Ar-
duino board which regulates the pulse modulation signal of each servomotor and the
Ethernet connectivity card used, Figure 3 illustrates the connection circuit designed
for the integration of the source to the Arduino control board and servomotors con-
nections of each arm.
Fig. 3. Control circuit.
iJOE ‒ Vol. 14, No. 1, 2018 189
Short Paper—Remote Lab for Robotics Applications
Figure 4 illustrates the implementation of the circuit and Arduino cards used, in the
area of the physical laboratory.
Figure 5 shows the robotic command interface of each of the degrees of freedom
per arm, which is located on the remote computer. For visual association of the user
two colors are used, red for arm one and blue for arm two. Shift bars are used for
greater agility in the variation of movement of the robotic arm, towards the left side
the angle decreases and towards the right it increases. The interface includes the field
of assignment of the IP address of the Arduino card in the remote area to be reached
by the Ethernet network. On the left side the visual feedback part is visible to the user
via the IP camera, the interface allows the IP address to be entered. The environment
of this application was implemented in C language under the Visual Studio software.
Fig. 4. Integration of circuits for robotic control.
Fig. 5. Remote monitoring and control tool
190 [Link]
Short Paper—Remote Lab for Robotics Applications
Figure 6 shows a flow diagram that guides the start-up logic of the remote applica-
tion. It must be initialized by doing the network connection, once done, it will initiate
the transfer of information from the remote application to the Arduino in the lab and
from the IP camera in the lab to the remote interface.
Fig. 6. Flow diagram of operation.
3 Results
To evaluate the remote working environment, a sequence of manipulation of the
robotic arms is done to perform a joint object grab task, this sequence consists in
manipulating the red robotic arm to pick up a didactic ball, take it to the end effector
of the blue robotic arm and it must grab it and move it to another location. Figure 7
illustrates the set sequence, for the case by 5 different users.
Fig. 7. Test sequence.
iJOE ‒ Vol. 14, No. 1, 2018 191
Short Paper—Remote Lab for Robotics Applications
Table 2 lists the results obtained when performing the sequence to the test users,
who were previously trained in the manipulation of the different degrees of freedom,
so that they had a certain level of familiarity with the remote environment and the
control of the robot. Accuracy is determined by tolerance values established as met-
rics in the execution of movements, for example, taking the ball properly and deliver-
ing it, without falling during the process, without exceeding the end position of the
desired end point, i.e. having to return to reposition some degree of freedom, etc.
Table 2. User Validations
Test 1 2 3 4 5
Accuracy (%) 92 88 82 78 89
Time (min) 3,57 4,12 4,46 4,51 4,38
From the results it can be observed that the task is performed by all users and, ac-
cording to the expertise of each one, it requires more or less execution time. Subjec-
tive factors such as caution are introduced, which at the time characterized the user 5
that with a precision of 89% handled the controls of each robot more carefully to
avoid passing a visual reference of displacement, according to the feedback of the
camera, which is why it took much longer, compared with user 2, who operated the
robots with almost the same precision.
There are some variations of latency in the communication with the IP camera, due
to the amount of information to be transmitted, proper to the video signal. These were
controlled by an iterative adjustment of the image capture resolution so that it would
not be unnecessarily reduced beyond the value according to a time delay noticeable
by sight.
4 Conclusions
A remote environment was designed for telecontrol of robotic agents, allowing the
proper manipulation of two robotic arms distant through network links. Because of
the design features, the remote platform is generic to the kind of user that can operate
it, given its simplicity. This design makes the platform functional to students of sec-
ondary or professional education.
With this virtual environment it can be developed and verified pre-program rou-
tines to the offline mode manipulator, facilitating the user the learning process in
telecontrol practices. Although kinematics is involved in the development of the con-
trol algorithm for each degree of freedom, including the spatial reference of the effec-
tor point to an object of interest would allow better metrics to be performed in preci-
sion tasks.
This type of environment allows a simple and inexpensive robotic telecontrol
learning interface, usable from anywhere with network connection, easily adaptable to
any type of robotic action sequence, such as telesurgery.
192 [Link]
Short Paper—Remote Lab for Robotics Applications
5 References
[1] M. Guo, P. Shi and H. Yu, "Development a feeding assistive robot for eating assist," 2017
2nd Asia-Pacific Conference on Intelligent Robot Systems (ACIRS), Wuhan, 2017, pp.
299-304. [Link]
[2] K. Swift-Spong, C. K. F. Wen, D. Spruijt-Metz and M. J. Matari!, "Comparing backstories
of a Socially Assistive Robot exercise buddy for adolescent youth," 2016 25th IEEE Inter-
national Symposium on Robot and Human Interactive Communication (RO-MAN), New
York, NY, 2016, pp. 1013-1018. [Link]
[3] M. Bowthorpe, M. Tavakoli, H. Becher and R. Howe, "Smith Predictor-Based Robot Con-
trol for Ultrasound-Guided Teleoperated Beating-Heart Surgery," in IEEE Journal of Bio-
medical and Health Informatics, vol. 18, no. 1, pp. 157-166, Jan. 2014.
[Link]
[4] He, Y. and Tan, M. (2017). Implementation of Air-conditioning Control Technology
Based on Wireless Sensor. International Journal of Online Engineering (iJOE), 13(07),
p.91. [Link]
[5] An, P. (2017). An Algorithm of Mobile Robot Node Location Based on Wireless Sensor
Network. International Journal of Online Engineering (iJOE), 13(05), p.4.
[Link]
[6] A"ıksoy, G. and Islek, D. (2017). The Impact of the Virtual Laboratory on Students’ Atti-
tude in a General Physics Laboratory. International Journal of Online Engineering (iJOE),
13(04), p.20. [Link]
[7] Gabriele, L., Marocco, D., Bertacchini, F., Pantano, P. and Bilotta, E. (2017). An Educa-
tional Robotics Lab to Investigate Cognitive Strategies and to Foster Learning in an Arts
and Humanities Course Degree. International Journal of Online Engineering (iJOE),
13(04), p.7. [Link]
[8] S. Nayak, P. Vakrani, A. Purohit and G. N. S. Prasanna, "Remote Triggered Lab for Ro-
botics: Architecture, Design and Implementation Challenges," 2014 IEEE Sixth Interna-
tional Conference on Technology for Education, Clappana, 2014, pp. 214-217.
[Link]
[9] I. Angulo et al., "RoboBlock: A remote lab for robotics and visual programming," 2017
4th Experiment@International Conference ([Link]'17), Faro, 2017, pp. 109-110.
[Link]
[10] L. Tobarra et al., "Low-cost remote laboratories for renewable energy in distance educa-
tion," 2014 11th International Conference on Remote Engineering and Virtual Instrumen-
tation (REV), Porto, 2014, pp. 106-111. [Link]
[11] R. Radhamani, N. Nizar, D. Kumar, B. Nair, K. Achuthan and S. Diwakar, "Low cost neu-
ro-inspired robots for sustainable laboratory education," 2016 International Conference on
Robotics and Automation for Humanitarian Applications (RAHA), Kollam, 2016, pp. 1-6.
[Link]
[12] M. Casini, D. Prattichizzo and A. Vicino, "A student control competition through a remote
robotics lab," in IEEE Control Systems, vol. 25, no. 1, pp. 56-59, Feb. 2005.
[Link]
[13] S. Kodagoda, A. Alempijevic, S. Huang, M. de la Villefromoy, M. Diponio and L. Cogar,
"Moving away from simulations: Innovative assessment of Mechatronic subjects using
remote laboratories," 2013 12th International Conference on Information Technology
Based Higher Education and Training (ITHET), Antalya, 2013, pp. 1-5.
[Link]
iJOE ‒ Vol. 14, No. 1, 2018 193
Short Paper—Remote Lab for Robotics Applications
6 Authors
Oscar F. Avilés S. was born in Bogotá, Colombia, in 1967. He received the Engi-
neer degree in Electronics and Specialization in Electronic Instrumentation from the
Antonio Nariño University – UAN - in 1995 and 2002, respectively. Master in Auto-
matic Production Systems - Technological University of Pereira - 2006 Colombia and
Ph.D. in Mechanical Engineering in Campinas State University – UNICAMP – Bra-
zil. Currently he is Professor in the Department of Mechatronics Engineering in Nue-
va Granada Military University - UMNG. He has experience in the areas of Electronic
Engineering, with emphasis on electronic instrumentation and control systems, acting
mainly on the following topics: robotics, control and biomechatronics.
Robinson Jiménez Moreno was born in Bogotá, Colombia, in 1978. He received
the Engineer degree in Electronics from the Francisco José de Caldas district Univer-
sity - UD - in 2002, respectively. Master in Industrial Automation from the Univer-
sidad Nacional de Colombia - 2012 and third year PhD student in Engineering of the
Francisco José de Caldas district University - UD. He is currently an Assistant Profes-
sor in the Mechatronics Engineering Program of the Nueva Granada Military Univer-
sity - UMNG. He has experience in the areas of Instrumentation and Electronic Con-
trol, acting mainly in: Robotics, Control, pattern recognition and image processing.
Mauricio F. Mauledoux M. was born in Bogotá, Colombia, in 1982. He received
the Engineer degree in Mechatronic engineering from the Nueva Granada Military
University, in 2005. In 2008 as a student of the Master in Information Technologies
and Intelligent Systems in the St. Petersburg State Polytechnic University, Russia, at
the automatic and intelligent distributed control department, he was promoted to a
Ph.D. In 2011 He received the Ph.D. degree in Mathematical models, numerical
methods and software systems (Red Diploma) from the St. Petersburg State Polytech-
nic University, Russia. In 2012, he joined the Department of Mechatronic Engineer-
ing, at Nueva Granada Military University, in Colombia, as an Assistant Professor.
His current research interests include Robotics, automatic control, Multi-agent Sys-
tems, Smart Grids, and Optimization.
Article submitted 06 September 2017. Published as resubmitted by the authors 13 November 2017
194 [Link]