Adaptability enhancement of multiple robots in dynamical environment

Abstract

This paper describes a scheme to enhance adaptability of the multiple robot system using prediction results. The main purpose of this paper is attempt to construct a prediction model, which actively incorporates knowledge of the environment and reflects reasoning results of the conditionally dependent behavior pattern, not only based on the conventional statistical analysis. The presented method will meet the demand and contribute to more efficient decision making in the dynamical environment. Also, some remaining problems are mentioned finally.