Analysis and Control of High-Speed Wheeled Mobile Robots

摘要

The overall objective of this research was to contribute to the development of control algorithms for autonomous high-speed vehicles moving in uncertain and/or off-road environments. The motivation behind this research objective was the need of 'intelligent' drivers that can be used to navigate and guide an autonomous vehicle in a high-threat environment, perform its assigned task, while at the same time minimizing exposure to potential threats and hazards. Given the previous overall objective, this research has focused on two areas: First, on the development of simple, albeit accurate, mathematical models for the complex behavior arising between the wheel tires and the ground. The study of tire friction dynamics is brought about by the need of high-speed operation of the vehicle. Our developed tire friction models are solidly based on first physical principles; they also capture transient dynamics which are important during high-speed and/or continuously changing driving conditions. Second, on the development of optimal, maximum-velocity and minimum-time driving maneuvers ('optimal driving primitives') under friction constraints. These results are also novel in the sense that the tire/ground nonlinearities (i.e., saturating friction) are explicitly accounted for. The numerical simulations show very realistic behavior of the vehicle trajectory, in the sense that sliding and skidding is often induced by the optimizer in order to achieve the control objective. This is not unlike the action of an expert human race driver who typically induces skidding and/or sliding to minimize time or maximize exit velocity from a corner.