Improved artificial potential field-based simultaneous forward search method for robot path planning in complex environment

Abstract

Our previous studies have discussed an effective improved artificial potential field based simultaneous forward search method (an Improved APF-based SIFORS method) in known and static environment. This paper presents the proposed Improved APF-based SIFORS method for autonomous mobile robot path planning in an environment including partially known or unknown, and dynamic complex environments. It can program a valid, feasible and shorter solution from the robot location to the target position. We redefine repulsive potential field and improve the wall-following method to solve intrinsic fatal problems of previous methods, and to address local minima and oscillation problems of the proposed method when a robot moves in a partially or entirely unknown environment. Using the latest five steps to assess the moving tendency of the robot, and combine the proposed wall-following method to an assistant robot to move out of local minima and oscillatory movements. To demonstrate the applications of our proposed method, we simulated paths planning for a local sensing range of a robot, a dynamic target, and moving obstacles in complex environments.