Beta Residuals: Improving Fault-Tolerant Control for Sensory Faults via Bayesian Inference and Precision Learning

TL;DR

This paper introduces a Bayesian fault-tolerant control method using beta residuals and precision learning, enabling implicit fault recovery and sensor fault detection within a unified framework, demonstrated through simulation results.

Contribution

It proposes a novel Bayesian FTC approach with beta residuals for fault detection, integrating fault recovery and detection without explicit fault isolation.

Findings

Beta residuals outperform competing methods in fault detection accuracy

The approach enables implicit fault recovery through stochastic control

Simulation results validate the effectiveness of the proposed method

Abstract

Model-based fault-tolerant control (FTC) often consists of two distinct steps: fault detection & isolation (FDI), and fault accommodation. In this work we investigate posing fault-tolerant control as a single Bayesian inference problem. Previous work showed that precision learning allows for stochastic FTC without an explicit fault detection step. While this leads to implicit fault recovery, information on sensor faults is not provided, which may be essential for triggering other impact-mitigation actions. In this paper, we introduce a precision-learning based Bayesian FTC approach and a novel beta residual for fault detection. Simulation results are presented, supporting the use of beta residual against competing approaches.