Optimal path planning for mobile robot using Preference based Evolutionary Qlearning algorithm

Abstract

An efficient Preference based Evolutionary Qlearning algorithm (PEQL) for mobile robot path planning method is proposed that finds optimal path between source to destination using an improved policy evaluation with adaptive reward and policy switching process, the proposed mobile robot path planning algorithm is unique and novel since it finds the optimal path using policy switching and better success index than the conventional methods. Keywords: mobile robot, path planning, adaptive reward, policy switching