Behavior control of multiple robots exploring unknown environment

Abstract

To control multi-robot exploring many targets in an unknown environment, the behavior control method should be set up. This paper presents six behavior control rules which contain searching rule along boundary, tracking target rule, avoiding repeatedly exploring rule, route choosing rule, maximizing explored area rule and obeying traffic regulation rule. The traffic regulations presented herein contain two solutions to avoid collision for robots' heading actions or cross actions, which can enhance the probability of robots' collision avoidance. The six behavior control rules are not pure independence, but with some correlations. When we establish the control model of multiple robots, the behavior correlation analysis is considered based on the linear weighted summation evaluation function, to enhance robots collaborative efficiency. The particle swarm optimal (PSO) algorithm is used to solve the behavior control model. The simulation result shows the behavior control method presented in the paper can effectively control the multiple robots disposing the multiple targets and avoiding collision.