MOBILE ROBOT OBSTACLE DETECTION AND ENVIRONMENT MODELING WITH SENSOR FUSION

Abstract

In this paper, modeling 2 D environment and road detection for mobile robot by fusing color and range image information are discussed. The environment model is constructed using multiresolution 2 D grid representation, which is proved to be a better solution to the tradeoff between accuracy and computation speed. The fusion algorithm is designed with a generalized Dempster Shafer theory of evidence ( DSTE ), which is efficient in dealing with dependent information. The multisensor system and fusion algorithm are tested in real environment and their effectiveness is proved.