Intelligent Adaptive Potential Field Motion Planning For Mobile Robots
Ahmad M. Alshorman, Mohammad A. Jaradat, Enas Ghabashneh, Mohammad Hamdan Garibeh · 2024
The mobile robot path planning and navigation in dynamic environments are of great significance for autonomous driving systems. This paper seeks to solve the current problems in artificial potential field (APF ) path planning, which are the problem of optimally targeting unreachable and the failure in complex environments. This paper proposes a hybrid intelligent adaptive potential field to tune the APF parameters for a dynamic environment. The adaptive potential field provides the robot with the needed forces for navigation through a safe path while moving in a dynamic environment. A fuzzy inference system, in essence, is used to determine the optimal attractive force parameter that must be provided to a robot for soft landing on a moving or fixed target. Furthermore, another fuzzy inference system is utilized to compute the repulsive force parameter that must be provided to a robot to repel it away from a moving or stationary obstacle. The results of comparison studies revealed a considerable improvement in failure rate and path length. Outcomes show that the suggested approach displays attractive features such as optimal path and zero failure rates in a robot environment with up to 20 static and dynamic obstacles, while the APF had a 16% failure rate and longer running time.