Fault diagnosis of full-drive AUV based on graph theory and recurrent neural network

This paper proposes a fault diagnosis method for multi-thruster Autonomous Underwater vehicles (AUV) based on graph theory and Recurrent Neural Networks (RNN). In the absence of feedback information from the thruster, the fault diagnosis can be effectively realized. First, the RNN is used to identify the parameters of the AUV model with constraints. Next, the model output is compared with the measured value to generate the fault residual sequence. Then, the residual sequence is converted into a directed graph and then feature extraction is performed to generate feature vectors. Finally, the feature vector is input into the Support Vector Machine (SVM) to realize fault location and classification. The verification results of multiple sea trials show that the average accuracy of the experiment can reach 90.46%.