Motion Planning for Tractor-Trailer System

Abstract

Motion planning for a tractor-trailer system is a challenging problem due to the nonholonomic constraints and highly nonlinear nature of the system. In this paper, we present a motion planning approach based on the Hybrid A* algorithm which produces smooth, collision-free, and kinematically feasible paths for a tractor-trailer system in a known environment. Kinematic model based on the on-axle hitching configuration of the tractor-trailer system is used for node expansion. Voronoi graph is used to compute sub-goals that guide the search towards the goal faster. A pure pursuit controller is employed to track the generated trajectory. The planning approach is evaluated over a number of Monte-Carlo simulation runs, and the experimental results indicate a perfect accuracy.