Neuroadaptive Output Tracking of Fully Autonomous Road Vehicles With an Observer

Abstract

Automated vehicle control systems are a key technology for intelligent vehicle highway systems (IVHSs). This paper presents an automated vehicle control algorithm for combined longitudinal and lateral motion control of highway vehicles, with special emphasis on front-wheel-steered four-wheel road vehicles. The controller is synthesized using an online neural-estimator-based control law that works in combination with a lateral velocity observer. The online adaptive neural-estimator-based design approach enables the controller to counteract for inherent model discrepancies, strong nonlinearities, and coupling effects. The neurocontrol approach can guarantee the uniform ultimate bounds (UUBs) of the tracking and observer errors and the bounds of the neural weights. The key design features are 1) inherent coupling effects will be taken into account as a result of combining of the two control issues, viz., lateral and longitudinal control; 2) rather ad hoc numerical approximations of lateral velocity will be avoided via a combined controller–observer design; and 3) closed-loop stability issues of the overall system will be established. The algorithm is validated via a formative mathematical analysis based on a Lyapunov approach and numerical simulations in the presence of parametric uncertainties, as well as severe and adverse driving conditions