Sensor-based behavior using a neural network for incremental learning in family mobile robot system

This paper proposes the way to apply the neural network for incremental learning to store the environmental information which autonomous children robots bring one by one. The parent robot manages the swarm of robots, obtains the information collected by their cooperative sensing, and makes the neural network memorize it. This information is used for path planning, by which the parent robot obtains the environment map when it is moving in an unknown environment. In addition, it is also shown that the parent robot can express the environment in the compact network and move as a result of learning and integrating the information on obstacles.<<ETX>>

[1]  Toshio Fukuda,et al.  Coordinative behavior by genetic algorithm and fuzzy in evolutionary multi-agent system , 1993, [1993] Proceedings IEEE International Conference on Robotics and Automation.

[2]  Fumihito Arai,et al.  A new neuron model for additional learning , 1992, [Proceedings 1992] IJCNN International Joint Conference on Neural Networks.

[3]  Toshio Fukuda,et al.  A neural network architecture for incremental learning , 1995, Neurocomputing.

[4]  Toshio Fukuda,et al.  Hierarchical intelligent control for robotic motion , 1994, IEEE Trans. Neural Networks.

[5]  Oussama Khatib,et al.  Real-Time Obstacle Avoidance for Manipulators and Mobile Robots , 1985, Autonomous Robot Vehicles.

[6]  Ren C. Luo,et al.  Multisensor integration and fusion in intelligent systems , 1989, IEEE Trans. Syst. Man Cybern..

[7]  Ronald C. Arkin,et al.  Behavior-Based Robot Navigation for Extended Domains , 1992, Adapt. Behav..

[8]  John Moody,et al.  Fast Learning in Networks of Locally-Tuned Processing Units , 1989, Neural Computation.

[9]  T Poggio,et al.  Regularization Algorithms for Learning That Are Equivalent to Multilayer Networks , 1990, Science.

[10]  Donald F. Specht,et al.  Probabilistic neural networks , 1990, Neural Networks.