Design of AUV height control based on Adaptive Neuro-Fuzzy Inference System

This paper studies the applications of Adaptive Neuro-Fuzzy Inference Systems(ANFIS) in the Autonomous Underwater Vehicle(AUV) height control system. A control method of ANFIS is mentioned after absorbing the advantages of fuzzy logic and neural network. According to the data fedback by sensors, this method can get the needed fuzzy membership functions and fuzzy rules by self-study, which composes the control system containing better performance than single neuro network system or single fuzzy system. Use the designed AUV height controller to complete the simulation experiments when AUV receives different height instructions on seafloor. The simulation experiments show that the controller has obvious effect in eliminating uncertainty, serious non-linearity and external interference. It possesses good dynamic performance and strong robustness.