Two-Stage Dichotomy Hybrid A* for Path Planning of Unmanned Mining Trucks in Large-Scale Open-Pit Mine
Yichen Zhang, Yafei Wang, Mingyu Wu, Zexing Li, Ruoyao Li · 2024
Planning a global reference path under final pose constraints is necessary for unmanned mining trucks to ensure the seamless execution of the loading process. However, irregular mountain shapes in large-scale mining areas lead to a substantial increase in the complexity of determining appropriate node expansion parameters of path planning. Furthermore, conventional methods which utilize fixed parameters in node expansion exhibit a dramatic rise in path length and computation time due to diverse obstacle distribution in openpit mines. To address this challenge, we propose a novel TwoStage Dichotomy Hybrid A* (TSD-HA*) to determine nearoptimal parameters in the node expansion process. Firstly, we generate convex polygonal obstacles through clustering on the grid map, thus reducing unnecessary node searches caused by U-shape obstacles. Subsequently, a two-stage dichotomy process is proposed to optimize steering angle and step length respectively, which yields near-optimal successor nodes and accelerates the node expansion process. To evaluate the effectiveness and computational efficiency of the proposed method, we conduct field experiments at an open-pit mine. The results demonstrate that our proposed method outperforms other conventional methods in terms of path length and computation time.