Particle Filter for Fault Diagnosis and Robust Navigation of Underwater Robot

Abstract

A particle filter (PF)-based robust navigation with fault diagnosis (FD) is designed for an underwater robot, where 10 failure modes of sensors and thrusters are considered. The nominal underwater robot and its anomaly are described by a switching-mode hidden Markov model. By extensively running a PF on the model, the FD and robust navigation are achieved. Closed-loop full-scale experimental results show that the proposed method is robust, can diagnose faults effectively, and can provide good state estimation even in cases where multiple faults occur. Comparing with other methods, the proposed method can diagnose all faults within a single structure, it can diagnose simultaneous faults, and it is easily implemented.

Navigation System Fault Diagnosis for Underwater Vehicle

“…By courtesy of Zhao Bo from NTNU [Zhao et al, 2012], [Zhao et al, 2014] recorded data from sea trials have been made available for analysis. Faults on sensors were induced and can be used to verify the methods discussed in this paper.…”

A Tutorial on Particle Filters for Fault Detection and Diagnosis of Hybrid Systems

“…Ref [17] designs a particle filter (PF) based robust navigation with fault diagnosis for an underwater robot. It uses a switching-mode hidden Markov chain models (HMM) to describe the navigation filter design for underwater robots and their anomalies.…”

Robust particle filter formulations with application to terrain‐aided navigation

“…The design and experimental tests of a PF‐based robust navigation system with fault diagnosis for a remotely operated underwater vehicle are described in . Given its ability to diagnose failures in Doppler‐based systems, including velocity biases and DVL dropouts, the method has high potential of application to terrain‐based navigation methods that rely on DVL for simultaneous dead‐reckoning and terrain data acquisition.…”