Abstract
This paper addresses the problem of mutual collision avoidance of mobile robots which work in an uncertain indoor environment. A local-information-based approach for collision-free-motion planning is proposed for the development of an autonomous and flexible multi-robot system. In order to reduce communication application, function of action observation is introduced into a robot. The robot is able to reason the space structure about its blind spot area by observing actions of other robots. It then plans its motions based on the results of space understanding and action observation. The effectiveness of this approach to collision avoidance is verified with simulation experiments.
