Online map building evolutionary algorithm for multi-agent mobile robots with odometric uncertainty

Abstract

An online map building evolutionary algorithm is proposed using multi-agent mobile robots with odometric uncertainty. The control algorithm for map building in each robot is identical and trained by an online evolutionary algorithm (EA). Each robot has configuration uncertainty which increases as it moves, and it perceives the surrounding environment information by the limited range sensors. It communicates with other robots and shares the information. The elementary behaviors are defined and they are used to build a map. EA is applied to the defined behavior set for optimizing the robot actions. To demonstrate the effectiveness of the proposed algorithm, computer simulations are conducted for various environments.