Improved Artificial Potential Field Method for Motion-Planning of Autonomous Vehicles
Pulkit Paliwal · 2023
Motion Planning is one of the primary tasks that needs to be taken care of for autonomous vehicles to function reliably. Given an environment to navigate, the vehicles must be able to navigate towards the desired goal location without colliding with any obstacles. Furthermore, one would want the vehicles to take the shortest route possible, in order to minimize the fuel consumption, and thus reduce the cost of operation. To find these optimal paths, several algorithms have been introduced over time. One such algorithm is the Artificial Potential Field Method, which considers an artificial field around the obstacles and goal, and by moving along the force given by the gradient of the resultant field, an reasonably optimal path is obtained. This paper discusses a variation of the Artificial Potential Field Method for path planning of autonomous vehicles which generates shorter paths than the ones generated using the original algorithm in a wide variety of environments. The simulation results presented reflect that the proposed algorithm reduces the path length by roughly 10% as compared to the path generated by the classical potential field method.