Minimum-time trajectory planning under dynamic constraints for a wheeled mobile robot with a trailer

This paper focuses on the minimum-time trajectory planning in a structured environment for a wheeled mobile robot with a trailer. The constraints which are accounted for include: obstacle avoidance, boundary conditions on the position/orientation of both the tractor and the trailer, maximum articulation angle between the two platforms, kinodynamic capacities of the actuators, and dynamic stability of the system. The proposed technique is based on the random profile approach, which uses a versatile master-slave optimization scheme. It is robust enough to handle different scenarios (free or imposed path) and different structures of the system (on axle or off-axle articulation). Simulation results show that it converges, in reasonable runtimes, towards good-quality solutions which are saturating the constraints. A preliminary experimental test suggests that the planned trajectories are realistically achievable.