Parking Path Planning Algorithm for Mine Rubber Tire Vehicle Based on Single-step Sampling

Addressing the path planning challenge for mine rubber tire vehicles in vertical, oblique, and parallel underground parking scenarios, this paper proposes a two-stage parking path planning algorithm based on single-step sampling optimization. Initially, the parking path is conceptualized in reverse, comprising the Entry phase and Adjustment phase, with the Runge-Kutta method employed for discretizing the path parameters. The algorithm employs the discrepancy between the vehicle’s current and target poses and the smoothness of the path as the cost function, transforming the parking path planning problem into an optimization control problem. Employing the single-step sampling optimization approach, the problem is solved by sampling control variables within the feasible domain, aiming to minimize the cost function associated with each state transition after discretization, thereby determining the sequence of control variables and iteratively constructing the parking path from the endpoint to the starting point. The results of simulation and real vehicle test prove that the proposed algorithm has strong adaptability to various parking scenarios, and can reduce the algorithm computation while ensuring the quality of path planning, which meets the real-time requirement in parking scenarios.