Continuous Curvature Parking Path Planning for Unmanned Mining Trucks Using Transition Curves and Model Predictive Control

Geometric methods based on Reeds–Shepp (RS) curves offer a practical approach for the parking path planning of unmanned mining truck, but discontinuous curvature can cause tire wear and road damage. To address this issue in mine scenario, a continuous curvature parking path planning method based on transition curve and model predictive control (MPC) is proposed for mine scenarios. Initially, according to the shovel position information issued by the cloud dispatching platform, a reference line is planned using RS curves. In order to mitigate the wear and tear of the tires and the damage to unstructured roads due to the in situ steering caused by the sudden change of the curvature, a transition curve consisting of clothoid–arc–clothoid that satisfies the kinematics of continuous vehicle steering is designed on the basis of RS curves to achieve the continuity of road curvature, which will contribute to the economy of tire and handling performance. The calculation of Fresnel integral involved by clothoid is simplified by using Chebyshev polynomial fitting method. Moreover, MPC is employed to re-plan an obstacle-avoidance path based on the reference line by designing a rational cost function. Finally, a simulation test platform is built considering the typical parking scenarios for unmanned mining trucks. The simulation results verify the effectiveness of the planning algorithm proposed in this article, and it shows the potential to reduce maintenance costs and improve mining efficiency.