Fault tolerant funnel control for uncertain linear systems

TL;DR

This paper introduces a low-complexity, model-free adaptive funnel control method for uncertain linear systems that ensures prescribed tracking performance and fault tolerance, demonstrated on an aircraft model.

Contribution

It develops a novel fault tolerant funnel controller using the time-varying Byrnes-Isidori form, allowing for systems with more inputs than outputs and actuator redundancy.

Findings

Achieves prescribed tracking error performance.

Ensures fault tolerance in uncertain linear systems.

Validated on a Boeing 737 aircraft model.

Abstract

We study adaptive fault tolerant tracking control for uncertain linear systems. Based on recent results in funnel control and the time-varying Byrnes-Isidori form, we develop a low-complexity model-free controller which achieves prescribed performance of the tracking error for any given sufficiently smooth reference signal. Within the considered system class, we allow for more inputs than outputs as long as a certain redundancy of the actuators is satisfied. An important role in the controller design is played by the controller weight matrix. This is a rectangular input transformation chosen such that in the resulting system the zero dynamics, which are assumed to be uniformly exponentially stable, are independent of the new input. We illustrate the fault tolerant funnel controller by an example of a linearized model for the lateral motion of a Boeing 737 aircraft.