Article

Article title COMPUTATIONAL SYNTHESIS OF CONTROL SYSTEM FOR GROUP OF ROBOTS GY SYMBOLIC REGRESSION
Authors A.I. Diveev, E.Yu. Shmalko
Section SECTION I. GROUP CONTROL ROBOTS
Month, Year 10, 2015 @en
Index UDC 004.896
DOI
Abstract We consider the problem of automatic synthesis of feedback controller for a robotic team. To synthesize a feedback control means to find a multidimentional nonlinear control function that depends on objects’ state. Such functions implemented to on-board computers will be able to produce control signals that transfer the control objects from the initial state to the terminal manifold with optimal value of quality criteria. The problem of control synthesis for the group of robots relates to the general problem of control synthesis and has additional dynamic constraints in the state space in order to avoid collisions of robots with each other in the process of movement towards the control objective. The review presents the three main approaches to constructing feedback control systems for groups of robots. These are methods that use navigation functions, that decompose the configuration space into cells and symbolic regression methods. In contrast to the first two approaches on multi robot coordination, symbolic regression methods allows synthesizing feedback controllers completely automatically, realizing the evolutionary search for the optimal control and its parameters in the space of possible solutions. Such an approach to the synthesis problem is versatile and not limited to the particular application problem. In the present paper we use such method of symbolic regression as the network operator method. The basic idea is to code any mathematical expression in the form of integer matrix which elements correspond to variables, parameters and elementary functions. The mathematical expression that meets the overall control objective with the best value of quality functional is searched by the genetic algorithm on the basis of the principle of small variations of basic solution. Our principle of small variations of the basic solution narrows the search space and reduces the computation time. When building a basic solution a specialist can use well-known analytical methods, such as analytical construction of aggregated controllers. The better the basic solution the faster a computer solves the problem. We validate our approach by presenting the results of applying the network operator method to the task of two mobile robots parking.

Download PDF

Keywords Control system synthesis; robotic team; dynamic constraints; symbolic regression; genetic algorithm.
References 1. Parker L. Alliance: An architecture for fault-tolerant multi-robot cooperation, Transactions on Robotics and Automation, 1998, No. 14 (2), pp. 220-240.
2. Arkin R., Balch T. Cooperative multiagent robotic systems, In D. Kortenkamp, R.P. Bonasso, and R. Murphy editors, Artificial Intelligence and Mobile Robots: Case Studies of Successful Robot Systems. MIT/AAAI Press, Cambridge, MA, 1998, pp. 277-296.
3. Arkin R., Balch T., Nitz E. Communication of behavioral state in multi-agent retrieval tasks, In IEEE International Conference on Robotics and Automation, Los Alamitos, CA, May 1993, pp. 588-594.
4. Matari_c M. Designing and understanding adaptive group behavior, Adaptive Behavior, 1995, No. 4, pp. 51-80.
5. Matari_c M. Issues and approaches in the design of collective autonomous agents, Robotics and Autonomous Systems, 1995, No. 16 (2-4), pp. 321-331.
6. Goldberg D., Matari_c M. Interference as a tool for designing and evaluating multi-robot controllers, In National Conference on Artificial Intelligence, Providence, Rhode Island, Jul 1997, pp. 637-642.
7. Lerman K., Martinoli A., Galstyan A. A review of probabilistic macroscopic models for swarm robotic systems, In Proceedings of the International Conference on the Simulation of Adaptive Behavior, Los Angeles, CA, Aug 2004, pp. 143-152.
8. Rimon E., Koditschek D. Exact robot navigation using artificial potential functions, IEEE Trans. Robotics and Automation, October 1992, Vol. 8, No. 5.
9. Conner D., Rizzi A., Choset H. Composition of local potential functions for global robot control and navigation, In Proc. IEEE Int. Conf. Intelligent Robots and Systems, Las Vegas, Nevada, October 2003.
10. Gennaro M. C. D., Jadbabaie A. Formation control for a cooperative multi-agent system using decentralized navigation functions, In Proceedings of the American Control Conference, Minneapolis, Minnesota, June 2006.
11. Conner D., Choset H., Rizzi A. Provably correct navigation and control for non-holonomic convex-bodied systems operating in cluttered environments, In Robotics: Science and Systems. Philadelphia, Pennsylvania: MIT Press, August 2006.
12. Habets L., J. van Schuppen A control problem for affine dynamical systems on a full-dimensional polytope, Automatica, January 2004, Vol. 40, No. 1.
13. Roszak B., Broucke M. E. Necessary and sufficient conditions for reachability on a simplex, Automatica, November 2006, Vol. 42, No. 11.
14. Lindemann S., Lavalle S. Computing smooth feedback plans over cylindrical algebraic decompositions, In Robotics: Science and Systems. Philadelphia, Pennsylvania: MIT Press, August 2006.
15. Loizou S.G., Kyriakopoulos K.J. Closed loop navigation for multiple holonomic vehicles, In Intl. Conf. on Robotics and Automation, Lausanne, Switzerland, October 2002.
16. Tanner H., Loizou S., Kyriakopoulos K. Nonholonomic navigation and control of cooperating mobile manipulators, IEEE Trans. Robotics and Automation, February 2003, Vol. 19, No. 1.
17. Loizou S.G., Kyriakopoulos K.J. Closed loop navigation for multiple holonomic vehicles, In Intl. Conf. on Robotics and Automation, Lausanne, Switzerland, October 2002.
18. Loizou S.G., Kyriakopoulos K. J. A feedback-based multiagent navigation framework, International Journal of Systems Science, 2006, Vol. 37, No. 6.
19. Lindemann S.R., LaValle S.M. Smoothly blending vector fields for global robot navigation, In IEEE Conference on Decision and Control, Seville, Spain, 2005.
20. Lindemann S.R., Hussein I.I., LaValle S.M. Real time feedback control for nonholonomic mobile robots with obstacles, In IEEE Conference on Decision and Control, 2006.
21. Koza J.R. Genetic Programming: On the Programming of Computers by Means of Natural Selection. Cambridge, Massachusetts, London, MA: MIT Press, 1992, 819 p.
22. O’Neill M., Ryan C. Grammatical Evolution. Evolutionary Automatic Programming in an Arbitrary Language. Kluwer Academic Publishers. 2002.
23. Zelinka I. Analytic programming by Means of Soma Algorithm, In Proceedings of 8th International Conference on Soft Computing Mendel02, Brno, Czech Republic, 2002, pp. 93-101.
24. Diveev A.I. A Numerical Method for Network Operator for Synthesis of a Control System with Uncertain Initial Values, Journal of Computer and Systems Sciences International, 2012, Vol. 51, No. 2, pp. 228-243.
25. Diveev A.I. Metod setevogo operatora [Method for network operator]. Moscow: VTs RAN, 2010, 178 p.
26. Lebedev B.K., Lebedev V.B., Lebedev O.B. Reshenie zadachi simvol'noy regressii metodami geneticheskogo poiska [The solution of the symbolic regression problem by genetic search methods], Izvestiya YuFU. Tekhnicheskie nauki [Izvestiya SFedU. Engineering Sciences], 2015, No. 2 (163), pp. 212-224.
27. Diveev A. I., Shmal'ko E. Metod simvol'noy regressii na osnove setevogo operatora v zadache sinteza upravleniya [The method of symbolic regression-based network operator in the problem of control synthesis], Sovremennye problemy nauki i obrazovaniya [Modern Problems of Science and Education], 2013, No. 3, pp. 76.
28. Kolesnikov A.A. Sinergeticheskie metody upravleniya slozhnymi sistemami: Teoriya sistemnogo sinteza [Synergetic control methods of complex systems: Theory of system synthesis]. Moscow: KomKniga, 2006, 240 p.
29. Kolesnikov A.A. Metod sinergeticheskogo sinteza sistemy upravleniya kolebaniyami «perevernutogo mayatnika na podvizhnoy telezhke» [Method of synergetics synthesis of “inverted pendulum on mobile cart” oscillation control system], Izvestiya YuFU. Tekhnicheskie nauki [Izvestiya SFedU. Engineering Sciences], 2011, No. 6 (119), pp. 110-117.
30. Oyama K., Nonaka K. Model Predictive Parking Control for Nonholonomic Vehicles using Time-State Control Form, In Proceedings of 2013 ECC European Control Conference, Zurich, Switzerland, 2013, pp. 458-465.

Comments are closed.