Research on Global Dynamic Path Planning Method Based on Improved A ∗ Algorithm

Abstract

Aiming at the optimal path and planning efficiency of global path planning for intelligent driving, this paper proposes a global dynamic path planning method based on improved $A^{\ast }$ algorithm. First, this method improves the heuristic function of the traditional $A^{\ast }$ algorithm to improve the efficiency of global path planning. Second, this method uses a path optimization strategy to make the global path smoother. Third, this method is combined with the dynamic window method to improve the real-time performance of the dynamic obstacle avoidance of the intelligent vehicle. Finally, the global dynamic path planning method of the proposed improved $A^{\ast }$ algorithm is verified through simulation experiments and real vehicle tests. In the simulation analysis, compared with the modified $A^{\ast }$ algorithm and the traditional $A^{\ast }$ algorithm, the method in this paper shortens the path distance by 2.5%∼3.0%, increases the efficiency by 10.3%∼13.6% and generates a smoother path. In the actual vehicle test, the vehicle can avoid dynamic obstacles in real time. Therefore, the method proposed in this paper can be applied on the intelligent vehicle platform. The path planning efficiency is high, and the dynamic obstacle avoidance is good in real time.