Author:
Yasuda Toshiyuki, ,Ohkura Kazuhiro,
Abstract
This paper describes an approach for controlling an autonomous homogeneous multi-robot system. An extremely important topic for this type of system is the design of an on-line autonomous behavior acquisition mechanism that is capable of developing cooperative roles as well as assigning them to a robot appropriately in a noisy embedded environment. Our approach applies reinforcement learning that adopts the Bayesian discrimination method for segmenting a continuous state space and a continuous action space simultaneously. In addition, a neural network is provided for predicting the average of the other robots’ postures at the next time step in order to stabilize the reinforcement learning environment. The proposed method is validated through computer simulations as well as our hand-made multi-robot system.
Publisher
Fuji Technology Press Ltd.
Subject
Electrical and Electronic Engineering,General Computer Science
Reference28 articles.
1. P. Stone, and R. S. Sutton, “Scaling Reinforcement Learning toward RoboCup Soccer,” Proc. of the 18th International Conference on Machine Learning, pp. 537-544, 2001.
2. F. Mondada, A. Guignard, M. Bonani, D. Floreano, D. Bär, and M. Lauria, “SWARM-BOT: From Concept to Implementation,” Proc. of IEEE/RSJ International Conference on Intelligent Robot and Systems, pp. 1626-1631, 2003.
3. B. Gerkey, and M. J. Matarić, “Pusher-Watcher: An Approach to Fault-Tolerant Tightly-Coupled Robot Coordination,” Proc. of IEEE International Conference on Robotics and Automation, pp. 464-469, 2002.
4. P. Stone, and M. Veloso, “Multiagent systems: survey from a machine learning perspective,” Autonomous Robots, 8(3): pp. 345-383, 2000.
5. M. Quinn, L. Smith, G. Mayley, and P. Husbands, “Evolving Team Behavior for Real Robots,” Proc. of EPSRC/BBSRC International Workshop on Biologically-Inspired Robotics, pp. 217-224, 2002.
Cited by
12 articles.
订阅此论文施引文献
订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献