Zhang, S.; Fan, L.; Gao, J., Pu, J., and Xu, K., 2018. Fault diagnosis of underwater vehicle and design of intelligent self-rescue system. In: Liu, Z.L. and Mi, C. (eds.), Advances in Sustainable Port and Ocean Engineering. Journal of Coastal Research, Special Issue No. 83, pp. 872–875. Coconut Creek (Florida), ISSN 0749-0208.

Due to the complexity of underwater environment, it is very necessary to study the fault diagnosis technology for underwater vehicle to perform underwater tasks safely, and it is also an important manifestation of its intelligence. According to the characteristics of propeller fault and its relationship with the motion state of underwater vehicle, the intelligent fault diagnosis and self-rescue system is designed for the underwater vehicle. Elman neural network is applied to establish the motion model of underwater vehicle, to obtain the fault information by comparing the model output and the actual measured value of motion. Experiments show that the design system can be used for on-line detection of propeller faults, to provide the basis for further fault tolerant control, and select the best measures for self-rescue, in order to ensure the vehicle to return safely in the condition of failure and even accomplish the task, which has important practical significance in underwater vehicle technology.