Robust tracking under uncertainties

Abstract

Robust tracking of vehicles under uncertain, noisy, and discontinuous positioning is a significant part of autonomous navigation in unknown environments. The article suggests two methods for track control, where the initial parameters of the online control are physically explainable, the resulting track as well as the control parameters are asymptotically converging and glitches in the localization are handled robustly. Practical experiments with landbound vehicles show the reliability and limitations of the method in various environments in setups for following simple attractors. Due to the physical meaning of the control parameters the adaptation to changed kinematic or dynamics is significantly simplified.