Motion planning through waypoints for a skid-steering mobile platform

This paper addresses the problem of motion planning through waypoints for the skid-steering mobile platform Rex. Proposed method is easily applicable to other robotics systems whose dynamic are represented by a control affine system with outputs. It is assumed that the waypoints are provided in the form of a sequence of points and velocities in the taskspace. The motion planning problem consists in finding control signals driving the platform in such a way that the platform passes through all the waypoints in a given time, preserving the order of the waypoints in the sequence. In order to solve this problem the Lifted Newton Method has been adopted, which can be regarded as a generalization of the well-known direct “multiple shooting” technique, widely applied to solving nonlinear optimization problems. A remarkable advantage of this method is a possibility of parallelization of the most computationally demanding parts of the motion planning algorithm. Performance of this algorithm has been illustrated by computer simulations.