A Predictor Based Linearization Approach for Local Stabilization of Nonlinear Systems With Input Delay

Abstract

In this paper, a predictor based linearization approach is presented to solve the local asymptotic stabilization (LAS) problem of time-invariant nonlinear systems with a large constant input delay. Under stabilizability and detectability of the linearized time-delay system, we prove that LAS is achievable by either predictor based state or output feedback controllers, which can be designed directly based on the linearization. The predictor based linearization technique developed in this research complements the work [1] and improves its result by removing the forward completeness condition and a number of technical requirements as well as by expanding the class of nonlinear systems with input delay to the general case. An example is given to demonstrate the application of the predictor based linearization technique.