Autonomous docking trajectory planning of heavy AGV based on slip compensation model

The track planning research of heavy AGV is the key to achieve accurate docking/undocking. The existing algorithms assume that the wheels are pure rolling, and do not consider the tire slip caused by the large inertia of heavy AGV in the actual situation. In order to realize that the trajectory planning is more in line with the requirements of real vehicles and make the AGV docking accuracy higher, the tire model of vehicle's transverse and longitudinal slip is derived, and the vehicle's incomplete constraint formula with slip motion is calculated to compensate the path error caused by slip; The path search uses the improved geometric target to inspire RRT. Based on the vehicle non-holonomic constraints, the line speed and wheel angle required for the RRT optimal step length are derived, and the search tree path satisfying the vehicle kinematics requirements is generated; In order to obtain the final trajectory, taking the curvature and smoothness of curvature as the objective functions, the conjugate gradient descent is used to optimize the redundant path solution, and the speed planning is allocated to generate the smooth trajectory. Finally, the rationality of the trajectory under the sliding model, the search efficiency of the improved RRT and the improvement of the solution quality by the optimization algorithm are compared through experiments.