Data-Driven Fault Detection of AUV Rudder System: A Mixture Model Approach

Based on data-driven and mixed models, this study proposes a fault detection method for autonomous underwater vehicle (AUV) rudder systems. The proposed can effectively detect faults in the absence of angle feedback from rudder. Considering parameter uncertainty AUV motion model resulting dynamics analysis method, we present identification based recurrent neural network (RNN). Prior to identification, singular value decomposition (SVD) was chosen denoise original sensor data as pretreatment step. provides more accurate predictions than recursive least squares (RLSs) single RNN. In order reduce influence errors prediction errors, adaptive threshold is mentioned analyzing errors. meantime, results were combined with qualitative force determine system’s diagnosis location. Experiments conducted at sea demonstrate feasibility effectiveness method.