Optimal Path Planning for Mobile Robot Based on Genetically Optimized Artificial Potential Field


This paper introduces a modified technique to find the shortest path between two points in known static environment for the mobile robot. The path planning in our proposal is based on the assumptions that; the robot is a small mass moving in two dimensions space with known static obstacles and subjected to an attractive force applied by the target as well as repulsive forces resultant from the obstacles. The combination of these forces moves the mass of robot directly toward the target in a manner that the mass of robot avoids all the obstacles on this way. The potential field is adapted (deformed (by manipulating potential field parameters according to static rules. The path of the mobile robot from start point to target point is optimized by choosing best values of the field parameters that give optimum form of potential field. The proposed genetic algorithm is used to search about these best values of field parameters. Simulation studies are carried out to verify and validate the effectiveness of the proposed method.