Delayed system control in presence of actuator saturation

Abstract

The paper is introducing a new design method for systems’ controllers with input delay and actuator saturations and focuses on how to force the system output to track a reference input not necessarily saturation-compatible. We propose a new norm based on the way we quantify tracking performance as a function of saturation errors found using the same norm. The newly defined norm is related to signal average power making possible to account for most common reference signals e.g. step, periodic. It is formally shown that, whatever the reference shape and amplitude, the achievable tracking quality is determined by a well defined reference tracking mismatch error. This latter depends on the reference rate and its compatibility with the actuator saturation constraint. In fact, asymptotic output-reference tracking is achieved in the presence of constraint-compatible step-like references.

Keywords

• Actuator saturation
• Reference tracking
• Dead-time systems
• Pole placement technique