Adaptive environment for developing and programming of mobile robots.
Pryanichnikov, Valentin E. ; Andreev, Victor Pavlovich ; Ivchenko, Valery 等
Abstract: This study considers the creation of a program complex on
the basis of a virtual machine with the operating system UNIX pre-installed
and a free software suite for elaborating and programming
applications for control systems of mobile technological robots. Being a
three-level software stack (virtual machine, operating system, and
development environment), the program complex is designed for students
to be used at laboratory classes of programming for AMUR mobile
technological robots within the course offered to universities, and
scientific-educational centers of the Russian Academy of Sciences. This
study analyses the existing program solutions at each level, sets
criteria for software selection (including the account for preparation
and skills of students) and gives recommendations on the selection of
specific software according to the original program.
Key words: educational and scientific stands, mobile robots,
mechatronics, sensorics, computer vision systems
1. INTRODUCTION
Traditional training of specialists in the integration of complex
technical systems with the help of conventional higher education methods
is hugely limited because students lack experience in building hardware
and software systems supporting the interaction between the
human-computer system and real objects of control. We managed to
implement a new model of the educational process: student tasks were
treated as scientific research to be conducted yearly and the results
obtained for each topic became the basis for new topics of research to
be conducted by another group. Such a "recursive" method of
research made it possible to achieve a suitable quality of research with
a successful methodical effect. Actual research in practice yielded a
technology of collective development of models and their implementation
into the control system for real mobile robots (MR).
If confined merely to mathematical models, the perception of
problems by students turned out to be incomplete. In view of this, we
developed a system of mechatronics tools and elements as educational and
scientific stands (mobile robots) designed for practical study, design,
and testing of complex technical systems as well as for performing
demonstration classes in the field of sensorics and intelligent systems.
The stands are suitable for testing and correcting students'
mistakes, which are "materialized" in the incorrect robot
operation, allowing one to promptly see the cause and cost of errors.
2. PROBLEMS OF EDUCATIONAL STANDS
Educational and scientific stands allow for a pictorial
demonstration and design of complex control systems for mechatronics
elements on the basis of analysis of the sensor-recorded situation. This
helps understand the nature of human interaction with complex technical
systems. The simulation of the interrelation between the elements of
technical systems and the use of individual and collective mapping
enables students to form a comprehensive view on the design, analysis
and synthesis, selection of criteria, and system evaluation. This is
achieved through the use of communication, interactive tools working in
conjunction with video cameras. The quality of results is satisfactory
due to the joint efforts of an educational institution (the Institute of
New Educational Technologies and Informatics of the Russian State
University for the Humanities), the Institute of Applied Mathematics of
the Russian Academy of Sciences and the International Laboratory
"Sensor".
The question arises: "Why it is the robotic and mechatronic
systems that are the best data?" The answer is:
1. Real mobile robots are psychologically easily projected on
humans--each student feels himself an expert in this field.
2. The universal computerization, including in the humanitarian
field, makes it necessary to perform in-depth training in the methods
for building intelligent software, which constitute one of the basic
problems of robotics.
3. Complex problems in mechatronics imply principally a creative
thinking without divestment or distancing from most topical problems of
modern production.
The main educational activities are carried out at the Institute of
New Educational Technologies and Informatics of the Russian State
University for the Humanities (in the Laboratory of problems of
informatics, mechatronics, and sensorics) as well as in the Moscow State
University of Equipment design and informatics. The education process is
characterized by active involvement of students and postgraduates into
the development and creation of units for prospective special mobile
caterpillar robots and simulators. This will enable the educational
problems and strong scientific projects to be joined into a unified
research process.
One can distinguish the following two stages:
1. Design of the module architecture with an onboard computer,
microprocessor control units, communication with intelligent sensors,
surveillance-scanning devices, power control switches, ultrasonic and TV
sensors, and software and microprogram support for navigation, object
identification, and management problems (Pryanichnikov et al., 2009);
2. Implementation of AMUR autonomous mobile educational robots,
which are used in the Far Eastern National Technological University and
the Institute of New Educational Technologies and Informatics together
with Festo stands.
3. TECHNICAL SUPPORT OF GROUP CONTROL
The problems of efficient information and group control arise in
education and in eliminating the consequences of emergency situations.
The paper (Andreev et al., 2010) presents our results on the creation of
a multi-element group of mobile robots operating as a unified
interrelated system capable of operating under combat conditions,
including the elimination of heavy disasters. Similar problems are posed
in the construction of robotic systems for space: development of methods
for group and remote control of autonomous systems based on network
technologies; self-diagnostics and self-learning; and organization of
the human--machine interface (Kuvshinov et al., 2006). The information
support of these activities is a key problem that can be efficiently
solved through the creation of a unified situational center and relevant
software and hardware tools for information analysis and management
support.
This formulation of the problem leads to the need for a multilevel
system of formation and exchange of information and control data
streams. In turn, this necessitates the construction of a mechanism that
can treat the interaction of a group of mobile robots and control
stations served concurrently by several operators as well as a
situational center as the top of hierarchy. A prerequisite of this
interaction is the organization of communication channels allowing the
unified command center of situation control to be allocated many
kilometers away from the work area. It should also be taken into account
that one normally employs MRs that should not be limited in the freedom
of movement. In addition, the bandwidth of communication channels must
be able to ensure the transfer of multi-stream video from computer
vision systems of mobile robots to both control stations and the
situation center. This problem we solved with the help or network
technologies; here, it is important that the elements of the complex
turn out to be mobile nodes of a local area network (LAN).
In solving these problems, our experience in the development of a
virtual distributed laboratory (Andreev et al., 2009) is useful. This
laboratory is designed for remote communication, including training of
professionals. The use of network technologies made it possible to
combine multiple MR, remote "satellites"-repeaters and remote
desks into a single computer network. The protection from unauthorized
access was organized through VPN-charmels, which were combined into a
single network of mobile robots located at the Institute of New
Educational Technologies and Informatics, in the building of the
Institute of Applied Mathematics of the Russian Academy of Sciences in
Moscow, and in the Institute for Automation and Control Processes of the
Far East Division of the Russian Academy of Sciences in Vladivostok
(IACP).
The investigations performed on this structure made it possible to
formulate the principles of construction and operation of educational
and research situational centers for the Federal Agency of Atomic Energy and the Russian Ministry for Civil Defense, Emergencies, and Elimination
of Consequences of Natural Disasters, which have widely used various
robotic systems. We have developed software that allows one not only to
remotely watch the "scene", but also to remotely control these
robots on the basis of a technology that avoids irregularities and
delays in communication channels. The computer network is based on the
TCP/IP stack of protocols because this is a currently widespread
practice. These kinds of networks have multiple technical
implementations, which allow different software and hardware units of
all elements of the complex to be combined into a single network. Then,
the problem is reduced to the creation of a LAN whose nodes are digital
IP-cameras and computing devices on-board a mobile robot, installed on
remote modules and repeaters, on the one hand, and on a computer on the
operator desk and computers of the situational center, on the other
hand; each with their own IP-addresses of a single network.
The studies aimed at the creation of a virtual distributed
laboratory resulted in a mechanism of control for multiple mobile robots
simultaneously. For this case, we developed a three-level control
system. The first level involves just the mobile robots with sensors
(ultrasonic, computer vision systems, etc.) and actuators. The second
level involves autonomous control systems for individual robots and
mobile robot operators. The link between robots was organized through a
Wi-Fi digital radiochannel. The third level involves systems of
top-level rules, controlled and modified by the commander, who
coordinate the operators as well. In this structure, the commander sees
the whole scene and receives video data from the computer vision system
of all units that are engaged in the operation of MR and remote units.
This level can be implemented (in constructing a computer network on the
basis of TCP/IP protocols) merely by including commander's computer
into the network and supply the system components by appropriate
software.
When using the above approach, the process of decision-making at
the level of the situation center using data of mobile robotic systems
that are processed by distributed local control centers becomes more
efficient and reasonable. The structure makes it possible to organize
global network connectivity at any point. This work was partially
supported by the Russian Foundation for Basic Research, project no.
10-07-00612a.
4. CONCLUSION
This study resulted in the creation of a cross-platform
distribution software of a virtual machine with the UNIX operating
system and a set of pre-installed open-source software for application
design and programming. The level of the virtual machine is implemented
using the Oracle Virtualbox; the level of the operating system is
implemented with the help of Ubuntu-11.04-desktop-i386; the development
environment is the Eclipse with the installed PyDev plugin and
Pythonrpyc and Python-pygame libraries. We have manged to limit the
volume of the distribution software to a DVD size (not more than 4 GB).
This provided students with a common development platform, which allowed
them to significantly accelerate the solution of given problems, to
improve the portability of codes and complementarity under the
conditions of teamwork. The development of AMUR educational and research
software for the IAPC and for the modernization of technological MR
BROKK-110D, BROKK-330 for the Russian Ministry for Civil Defense,
Emergencies, and Elimination of Consequences of Natural Disasters showed
that the proposed approach is efficient and reduces the time needed for
creation and debugging of program complexes several times as compared to
traditional approaches (Pryanichnikov et aL, 2009).
5. REFERENCES
Andreev V.P, Kirsanov K.B., Levinski B.M., Pryanichnikov V.E. [et
al.] (2009). Problems of the construction of a network of Internet labs
using autonomous mobile robots AMUR. Information-Measuring and Control
Systems, Moscow: radiotekhnika, 2009, vol. 7, No. 6, pp.124-131
Andreev V., Pryanichnikov V., Prysev E. (2010). Multi-access
control of distributed mobile robotic systems on the base of networking
technologies, Annals of DAAAM for 2010&Proceedings of the 21st
International DAAAM Symposium, ISSN 1726-9679, pp. 15-16, Zadar, Croatia
Katalinic, B., Kordic, V. (2002) Concept of Design and Scheduling
of Self-Organizing Complex Flexible Assembly System, Proceedings of the
4th International Workshop on Emergent Synthesis--IWES 02, ISBN 961-4238-49-3, pp. 89-96, March 12-13, 2002, Kobe, Japan
Kuvshinov S.V., Pryanichnikov V.E., Artemenko O.L., Levinski B.M.
(2006). Development of the elements of an educational and scientific
environment for problems of mechatronics and sensorics.
Information-Measuring and Control Systems, No. 1-2, vol. 4, 2006,
Moscow: Radiotekhnika, pp. 170-175
Pryanichnikov V., Andreev V., Prysev E. (2009). Computer vision and
control for special mobile robots, Annals of DAAAM for 2009 &
Proceedings of the 20th International DAAAM Symposium "Intelligent
Manufacturing & Automation" 25-28th November 2009. ISSN
1726-9679, pp. 1857-1858, Vienna, Austria
