A Decoding Approach to Fault Tolerant Control of Linear Systems with Quantized Disturbance Input

Abstract: The aim of this paper is to propose an alternative method to solve a Fault Tolerant Control problem. The model is a linear system affected by a disturbance term: this represents a large class of technological faulty processes. The goal is to make the system able to tolerate the undesired perturbation, i.e., to remove or at least reduce its negative effects; such a task is performed in three steps: the detection of the fault, its identification and the consequent process recovery. When the disturbance function is known to be \emph{quantized} over a finite number of levels, the detection can be successfully executed by a recursive \emph{decoding} algorithm, arising from Information and Coding Theory and suitably adapted to the control framework. This technique is analyzed and tested in a flight control issue; both theoretical considerations and simulations are reported.