A SINGULARITY-BASED APPROACH TO IMPROVE MOBILE ROBOT MOTION PLANNING

L. Gracia∗ and J. Tornero∗

Keywords

Path planning, trajectory planning, wheeled mobile robots

Abstract

This paper presents a singularity-based method that complements path-planning and motion-planning techniques (e.g., tree graphs, roadmaps, etc.) in order to select the optimal collision-free path or trajectory among several possible solutions. Slip motions, infinite estimation error and impossible control actions are avoided escaping from singularities. High amplification of wheel velocity errors and high wheel velocity values are also avoided by moving far from the singularity. To illustrate the applications of the proposed approach, the method is used to plan paths for an industrial forklift in a simulated environment. Several results are shown for the proposed optimality criterion, which are compared to those obtained with other classical criteria, such as shortest path and minimum energy.

Important Links:



Go Back