Information fusion fault diagnosis method for unmanned underwater vehicle thrusters

Abstract

A novel thruster fault diagnosis method for open-frame unmanned underwater vehicle is presented in the study. An credit assignment-based fuzzy cerebellar model articulation controller (FCA-CMAC) neural network information fusion model is used to realise the fault identification for thruster continuous and uncertain jammed fault situations. Information inputs for the fusion model are yaw rate and the control signal for the underwater vehicles; the information output of the FCA-CMAC fusion model is the corresponding jammed fault parameter s, which indicates the degree of the fault. To illustrate the effectiveness of the proposed method, a pool experiment under uncertain continuous fault conditions is presented in this study.