Hierarchical Motion Planning for Multisteering Tractor–Trailer Mobile Robots with On-Axle Hitching

Abstract

Normally, due to the strongly coupled kinematics and under-actuated property, motion planning for a multisteering tractor-trailer mobile robot (MSTTMR) is a complex problem. In this paper, an MSTTMR system with on-axle hitching is considered. Motion planning for it is divided into three independent subproblems: planning a collision-free geometric path for the tractor, transforming the geometric path into a motion trajectory of the tractor, as well as generating motion trajectories for the trailers, and then they are solved by means of a three-level hierarchical method. At the highest level, a geometric path in the form of a cubic spline curve is planned for the tractor. To consider obstacle avoidance and satisfy constraints on the control inputs of the tractor, the path parameters are optimized by an improved quantum-behaved particle swarm optimization algorithm. At the middle level, the planned path is transformed into a feasible motion trajectory of the tractor. At the lowest level, to guide all the trailers to track the planned path, motion planning for the trailers is achieved by the Gauss pseudospectral method (GPM). As a result, both the tractor and trailers can simultaneously move along the same path, and turning maneuvers in narrow and cluttered spaces can successfully be achieved. Simulation and experimental results are presented to show the effectiveness of the proposed approach.