Path planning algorithm for articulated loader based on bidirectional Dubins curve

Abstract Wheel loaders are often used for material transportation in areas such as mines and construction sites, and their articulated motion characteristics bring great challenges to their path planning and trajectory tracking. Aiming at the problems of long planning path and low tracking accuracy in the path planning of articulated loaders, this paper proposes an improved hybrid A* algorithm based on bidirectional Dubins curve to realize the path planning and trajectory tracking in the process of automatic material transportation of loaders. Firstly, the kinematics model is constructed based on the vehicle size, and the bidirectional Dubins curve combined with hybrid A* algorithm is introduced to improve the forward and backward switching mechanism of the loader, and the path is smoothed by factor graph. Then, the kinematics model is analyzed, and the mapping relationship between navigation point speed and vehicle speed and articulation angle is obtained, and the vehicle state is obtained by sensor feedback. Finally, the vehicle speed and articulation angle are solved by graph optimization. The simulation and field test verify its performance, and the path length is shortened by 17.73% compared with that before optimization. The research results show that the proposed algorithm can accurately control the wheel loader from the initial position to the target position, realize the stable work of the whole vehicle in the work area, and provide technical support for the loader's autonomous operation.