基于自适应人工势场的自动驾驶矿用卡车路径规划系统研究

Research on Path Planning System for Autonomous Mining Trucks Based on Adaptive Artificial Potential Field

实现矿用卡车自主路径规划对保障作业安全、提高作业效率及改善工作环境具有重要意义。针对矿区独特的地形环境,构建了相应的人工势场对路径进行评价,并结合模型预测控制,利用其预测能力提高规划路径的前瞻性。同时,将不同的障碍分为不可通过的障碍、可通过的障碍和道路边界,利用模糊推理规则根据具体环境对不同的障碍进行权重分配,提高矿用卡车路径规划系统对矿区特殊环境的适应能力。最后,在Carsim-Matlab联合仿真平台对多种驾驶情景进行了测试,测试结果表明,与对比方法相比,基于自适应人工势场的路径规划方法能够根据驾驶场景发生碰撞的紧急程度实施路径规划,从而提高了对矿区不同行驶环境的适应能力,矿用卡车始终与障碍保持2 m以上的距离,实现了无碰撞的自动驾驶。

Implementing autonomous path planning for mining trucks is of great significance for ensuring work safety,enhancing work efficiency,and improving the work environment.According to the unique terrain environment of the mining area,the corresponding artificial potential field was constructed to evaluate the path,and combined with model predictive control,its predictive ability was used to improve the forward-looking of the planned path.At the same time,different obstacles were divided into non passable obstacles,passable obstacles and road boundaries,and fuzzy inference rules were used to assign weights to different obstacles according to the specific environment,so as to improve the adaptability of the mining truck path planning system to the special environment of the mining area.Finally,multiple driving scenarios were tested on the Carsim Matlab joint simulation platform.The test results show that,compared with the comparative method,the path planning method based on adaptive artificial potential field can implement path planning according to the urgency of collision in the driving scenario,thereby improving the adaptability to different driving environments in the mining area.The mining truck model always maintains a distance of more than 2meters from obstacles,realizing collision free autonomous driving.