An Optimization-Based Path Planning Approach for Autonomous Vehicles Using the DynEFWA-Artificial Potential Field
Hongcai Li, Wenjie Liu, Chao Yang, Weida Wang, Tianqi Qie, Changle Xiang
Abstract
With the rapid development of autonomous driving technology, collision avoidance has become a research hotspot since it has the potential to increase safety. To obtain a collision-free path, the artificial potential field (APF) is widely used as a path planning method. APF is capable of establishing functional relationships between the vehicle and surrounding objects. However, the function features of the traditional APF method can cause autonomous vehicles to fall into the local minimum, and the generated zigzag path may be difficult to follow. Motivated by these challenges, this paper proposes a real-time path planning method for autonomous vehicles using the dynamic enhanced firework algorithm-APF. Firstly, to improve the safety and smoothness of the planned path by the traditional APF method, the constraints of the vehicle dynamics and different types of obstacles are taken into consideration. Secondly, an optimization problem is formulated to find an optimal path with the least cost in the driving area. Finally, the proposed method is verified with both a simulation and a hardware-in-loop test environment. The results show that the studied autonomous vehicle successfully avoids obstacles and arrives at the goal position by using the proposed path-planning method, and the path smoothness is improved.