A Real-Time Path Planning Algorithm Based on the Markov Decision Process in a Dynamic Environment for Wheeled Mobile Robots

A real-time path planning algorithm based on the Markov decision process (MDP) is proposed in this paper. This can be used dynamic environments to guide wheeled mobile robot goal. Two phases (the utility update phase and policy phase) constitute of entire system. In phase, value updated information from observable environment. Obstacles walls reduce value, pushing agents away these impassable areas. The goal constant always only largest. update, a series policies obtained by strategy maximizing its long-term total reward, will eventually form towards goal, regardless where agent located. simulations experiments show that it takes longer find first beginning due large changes but once planned, requires small amount time cost respond environmental changes. Therefore, has an advantage obstacles move unpredictable ways.