1A2-D10 A Basic Study on Cooperative Behaviors Acquisition of Multiple Robots(Evolution and Learning for Robotics(2))

Abstract

In this study, we treat the cooperative agents systems in which there are multiple autonomous mobile robots, and the 3-Dimensional balancing problem is given. This problem is an example of corresponding tasks to find the appropriate movements for multiple mobile robots. Each robot agent on a flat board keeps being balanced state by using evolutionary robotics approach. The behaviors of mobile robots are calculated by corresponding neuro-controller which consists of artificial neural networks. Particle swarm optimization method is applied to evolve weight values of the neuro-controller. Numerical simulations are carried out by a simulator constructed based on 3-D physical engine PhysX. As a result of simulations, usefulness of our approach is clarified.