Resumen
Path planning is one of the key technologies for unmanned driving. However, global paths are unable to avoid unknown obstacles, while local paths tend to fall into local optimality. To solve the problem of unsmooth and inefficient paths on multi-angle roads in a park which cannot avoid unknown obstacles, we designed a new fusion algorithm based on the improved A* and Open_Planner algorithms (A-OP). In order to make the global route smoother and more efficient, we first extracted the key points of the A* algorithm and improved the node search structure using heap sorting, and then improved the smoothness of the path using the minimum snap method; secondly, we extracted the key points of the A* algorithm as intermediate nodes in the planning of the Open_Planner algorithm, and used the A-OP algorithm to implement the path planning of the unmanned sweeper. The simulation results show that the improved A* algorithm significantly improved the planning efficiency, the nodes are less computed and the path is smoother. The fused A-OP algorithm not only accomplished global planning effectively, but also avoided unknown obstacles in the path.