| Abstract: | The automated shovel loading of mining vehicles is one of the core operations of intelligent mining in open pit mines. However, in some complex mining scenarios, the traditional path planning methods for autonomous driving are time-consuming and the output path has many unnecessary curvature changes without sufficient consideration of safety, which leads to the reduction of operational efficiency and the safety factor. Considering the complex operational scenarios for automated mining vehicles and aiming at the high
real-time and safety requirements of path planning, this paper proposes a path planning method based on the guided variable-step-length hybrid A* algorithm. Initially established on the Voronoi diagram, the mine road network is obtained and the key points are extracted as directional guidance in order to improve the planning efficiency and prevent the vehicle from falling into local optimum at U-shaped obstacles. And then the adaptive variable-step-length algorithm is introduced and the heuristic function is reset to further improve the planning efficiency and path safety. Finally the effectiveness of the algorithm is verified by the application in the real mine scenes. The experimental results show that the path planning method proposed in this paper meets the requirements of complex scenarios for mining vehicles. The planning time is reduced by 68% compared with the original guided algorithm, the average path-to-obstacle distance is increased by 11% and the number of path curvature changes is reduced by 45%, which significantly show the improvement in
computational efficiency and path quality. |
