Potential Field Path Planning and Potential Field Path Tracking

Abstract

In robotics studies, problems can be given as mapping, location estimation, exploration and navigation. Navigation problem; It can be subdivided into path planning, path tracking and obstacle avoidance. In the study, in the Gazebo simulation environment, a 4-wheel robot and a laser distance sensor were used. 2D mapping, path planning and path following with respect to the 2D map are implemented on Robot Operating System (ROS). Potential-based path planning was compared with the A* algorithm. Potential-based path tracking was compared with the dynamic window approach (DWA). In tests conducted in the Gazebo simulation environment, it was observed that the potential-based approach yielded successful results in path planning and path tracking.