Smooth Local Planning Incorporating Steering Constraints

Individually controllable and steerable wheels provide vehicles with greater flexibility and efficiency. A key challenge in the context of such vehicles lies in the effective planning of the motion commands in order to properly deal with the potentially large required changes in the steering angles and to properly incorporate the given motion constraints. Especially large vehicles often perform large steering angle changes only while standing to prevent physical damage to the robot. In this paper we introduce a novel planner for vehicles with individually controllable and steerable wheels and two novel methods to incorporate steering constraints. The latter are designed to provide smooth changes of the instantaneous center of rotation (ICR). Extensive experiments in simulation and on the real robot reveal that our methods allow for an effective planning of steering commands, reduce the travel time and additionally enable the vehicle to accurately follow a given path.