An autonomous underwater vehicle (AUV) should have the ability of adapting the complexity and unpredictability of the marine environment, which means that the technology of AUV’s fault diagnosis is very significant, especially the part of thrusters. In order to make it possible, one fault diagnosis strategy of AUV’s thrusters is proposed, which is based on the support vector machine (SVM). SVM has many unique advantages in solving small-sample, nonlinear and high dimensional problems. In this paper, different character signal is inputted SVM to train and test it. The simulation results show that the fault diagnosis of AUV’s thrusters based on offline SVM can classify the fault styles successfully, which proves its feasibility and effectiveness. This method offers a new way to solve the fault diagnosis of AUVs.
Estimation of Volumetric Oxygen Concentration in a Marine Environment with an Autonomous Underwater Vehicle
