The Influence of the Model Discrepancies in the Jacobian Motion Planning Algorithm for Nonholonomic Systems

The Jacobian motion planning of a nonholonomic system, derived by means of the Endogenous Configuration Space Approach, relies on a complete knowledge of a robot's model. The aim of this paper is to identify and investigate the influence of insufficient knowledge of the model's parameter values on the quality of the Jacobian motion planning algorithm results. To address the problem, we propose an Iterative Learning Control scheme endowed with the Endogenous Configuration Space Approach. The efficiency of the presented proposition is represented by the numerical experiments.