Journal of the Korean Institute of Intelligent Systems (한국지능시스템학회논문지)

Korean Institute of Intelligent Systems (한국지능시스템학회)
권호 표지
DOI QR코드

DOI QR Code

A Rule-based Integration of Neural Network Modules based on Cellular Automata for Sensory-Motor Controller

센서-모터 제어기를 위한 셀룰라 오토마타 기반 신경망 모듈의 규칙기반 결합

  • 김경중 (연세대학교 검퓨터과학과) ;
  • 송금범 (연세대학교 검퓨터과학과) ;
  • 조성배 (연세대학교 검퓨터과학과)
  • Published : 2002.02.01
Download PDF
⟨ Previous Next ⟩

Abstract

There are some difficulties to construct a sensory-motor controller for an autonomous mobile robot such as coordinating the mechanics and control system parts of the robot, and managing interaction with external environments. In previous research, we evolve the CAM-Brain, neural networks based on cellular automata, to control an autonomous mobile robot. In this paper, we propose the method of combining multi-modules evolved to do simple behavior in order to making more sophisticated behaviors because the controller composed of one neural network module is difficult to make complex behaviors. In experimental results, we can get the controller adapting to more complex environments by combining CAM-Brain modules evolved to do simple behavior by rule-based approach.

자율이동로봇의 센서-모터 제어기를 구축하는데 있어 로봇의 기계적인 부분과 제어기 부분을 조화시키는 것이나 외부환경과 로봇의 상호작용을 처리하는 것 등이 가장 큰 문제점이다. 이러한 문제점들을 해결하기 위해서 진화적 접근방법이 많이 사용되고 있다. 이전 연구에서는 이러한 연구선상에서 셀룰라 오토마타 기반 신경망인 CAM-Brain을 이동로봇 제어기로 진화시켰다. 그러나, 하나의 모듈로 이루어진 제어기로는 복잡한 행동을 하도록 만들기 어렵기 때문에 본 논문에서는 하위 수준의 간단한 행동을 하도록 진화된 모듈들을 결합하여 보다 상위 수준의 복잡한 행동을 하도록 하는 다중 모듈 결합방법을 제안한다. 실험결과, 간단한 행동들을 하도록 진화된 CAM-Brain 모듈들을 규칙기반 방법으로 결합하여 주어진 좀더 환경에 적응할 수 있는 제어기를 얻을 수 있었다.

Keywords

References

  1. I. Harvey, P. Husbands and D. Cliff, "Issues in evolutionary robotics," Proc. of the Second lnt. Conf. on Simulation of Adaptive Behavior (SAB92), pp. 364-373, Cambridge MA, 1993.
  2. S. Nolfi, D. Floreano, O. Miglino and F. Mondada, "How to evolve autonomous robots: different approaches in evolutionary robotics," Proc. of the Int. Conf. on Artificial Life IV. pp. 190-197, 1994.
  3. 송금범, 조성배, "셀룰라 오토마타 기반 신경망의 진화 및 행동분석," 한국정보과학회 논문지, 제 26권, 제 4호, pp.453-461, 1999.
  4. H. de Garis, "CAM-Brain: ATR's billion neuron artificial brain project," Proc. of Int. Conf. on Evolutionary Computation, pp. 886-891, Nagoya, Japan, 1996.
  5. F. Gers, H. de Garis and M. Korkin, "CoDi-1Bit: A simplified cellular automata based neuron model," Proc. of Artificial Evolution Conf., Nimes, France, 1997.
  6. M. Korkin, H. de Garis, F. Gers and H. Hemmi, "CBM(CAM-BRAIN" MACHINE): A hardware tool which evolves a neural net module in a fraction of a second and runs a million neuron artificial brain in real time," Proc. of Genetic Programming Conf., Stanford, USA, 1997.
  7. N.E. Nawa, H. de Garis, F. Gers and M. Korkin, "ATR's CAM-Brain Machine (CBM) simulation results and representation issues," Proc. of Genetic Programming Conf., USA, 1998.
  8. H. de Garis and 조성배, "CAM-Brain:ATR의 인공두뇌 프로젝트," 정보과학회지, 제17권, 제5호, pp.20-26, 1999.
  9. W. Banzhaf, P. Nordin and M. Olmer, "Generating adaptive behavior using function regression within genetic programming and a real robot," Int. Conf. on Genetic Programming, pp. 35-43, 1997.
  10. M. J. Mataric, "Designing and tu1derstanding adaptive group behavior," Adaptive Behavior, vol. 4, no. 1, pp. 51-80, 1995. https://doi.org/10.1177/105971239500400104
  11. K-Team, Khepera Simulator Version 5.02 User Manual, 1999.
  12. D.T. Cliff, P. Husbands and L Harvey, "Analysis of evolved sensory-motor controllers," Proc. of the 2nd European Conference on Artificial Life (ECAL93), pp. 192-204, 1993.
  13. D.T. Cliff, I. Harvey and P. Husbands, "Incremental evolution of neural network architectures for adaptive behaviour," Proc. of the First European Symposium on Artificial Neural Networks, pp. 39-44, Brussels, Belgium, 1993.
  14. I. Harvey, P. Husbands and D. Cliff, "Seeing the light: Artificial evolution, real vision," Proc. of 3rd lnt. Conf. on Simulation of Adaptive Behavior '94, pp. 392-401, 1994.
  15. D. Floreano and F. Mondada, "Evolution of homing navigation in a real mobile robot," IEEE Trans. Systems, Man, and Cybernetics, vol. 26, no. 3, pp. 396-407, 1996. https://doi.org/10.1109/3477.499791
  16. R. Walker and O. Miglino, "Simulating exploratory behavior in evolving artificial neural networks," Proc. of the Genetic and Evolutionary Computation Conf., pp. 1422-1428, Orlando, USA, 1999.
  17. J.R Koza, Genetic Programming: On The Programming of Computers by Means of Natural Selection, MIT Press, Cambridge, Massachusetts, 1992.
  18. J.R. Koza, and J.P. Rice, "Automatic programming of robots using genetic programming," Proc. of the 10th National Conf. on Artificial Intelligence, pp. 194-201, 1992.
  19. M. Ebner, "Evolution of a control architecture for a mobile robot," Proc. of the 2nd Int. Conf. on Evolvable Systems: From Biology to Hardware, pp. 303-310, Lausanne, Switzerland, 1998.
  20. M. Ebner and A. Zell, "Evolving a behavior-based control architecture - From simulations to the real world," Proc. of the Genetic and Evolutionary Computation Conf., pp. 1009-1014, Orlando, USA, 1999.
  21. P. Nordin and W. Banzhaf, "Real time control of a Khepera robot using genetic programming," Control and Cybernetics, Vol. 26, No. 3, pp. 533-561, 1997.
  22. P. Nordin, W. Banzhaf and M Brameier, "Evolution of a world model for a miniature Robot using genetic programming," Robotics and Autonomous Systems, Vol. 25, pp. 105-116, 1998. https://doi.org/10.1016/S0921-8890(98)00004-9
  23. S.B. Cho and S.I. Lee, "Mobile robot learning by evolution of fuzzy controller," Journal of Intelligent and Fuzzy Systems, Vol. 6, pp. 91-97, 1997.
  24. L.A. Zadeh, "Fuzzy sets," Information and Control, Vol. 8, pp. 338-353, 1965. https://doi.org/10.1016/S0019-9958(65)90241-X

Cited by

  1. A neural network based multi-classifier system for gene identification in DNA sequences vol.14, pp.2, 2005, https://doi.org/10.1007/s00521-004-0447-7