A PATH PLANNING APPROACH TO MOBILE ROBOT UNDER DYNAMIC ENVIRONMENT

Abstract

This paper presents a new approach to path planning under dynamic environment. MAKLINK graph is used to model the workspace. The whole system includes two parts: the global model decides the global path planning by genetic algorithm, and the local model develops the global path. Three primitive behaviors called global path following, obstacle avoidance and head for goal behaviors are used for the local path planning. Particularly, we introduce reinforcement algorithm for obstacle avoidance behavior. The results of simulation experiment indicate the effectiveness of the proposed method.