Motion analysis and feedforward control of a tail-slide vehicle

Conventional motion planners for wheeled vehicles often attempt to prevent wheel slipping or even assume no-slipping condition. However, in some practical situations slipping can be useful. The purpose of this work is construct a motion planning strategy for tail-slide vehicles where the rear wheel is slipping. First we develop a dynamic model of a four-wheeled vehicle by combining an extended 2D-model of the Coulomb friction and the nonholonomic rolling constraints. We then show that, compare to model ignoring sliding, the turning radius of the vehicle can be made smaller by using the tail-slide. After revealing the realizability condition of the tail-slide, we construct a motion planning strategy for steady-state cornering by combining the tail-slide and nonholonomic motion maneuvers. The feasibility of the motion planning strategy is tested under simulation.