Global Stabilization of a Class of Uncertain Systems with Saturated Adaptive Robust Control

In this paper, a class of saturated adaptive robust control (SARC) laws are developed for nonlinear systems in the \chain-of-integrator" form with both parametric uncertainties and non-repeatable uncertainties. A guaranteed transient performance and nal tracking accuracy is achieved in general. Furthermore, asymptotic output tracking is also achievable provided that the system undergoes parametric uncertainties only. Discontinuous projection method is used in the adaptation law for a controlled learning. Given the saturation limits of control authority, certain criteria are obtained to predict the achievable high-performance working range of the closed-loop system by taking into account the order of system, the bounds of both parametric uncertainties and non-repeatable nonlinearities, and the required performance, such as reaching time. At the same time, these criteria can also be used in the trajectory planning to obtain realizable desired trajectories. Consequently, an integrated design of achievable desired trajectory and control law may be achieved. The proposed SARC is then applied to the control of a linear motor drive system, and an excellent output tracking performance is obtained in experiments.