Sliding mode control of ROV based on RBF neural networks adaptive learning

This paper deals with a variable structure sliding mode control of ROV (remotely operated vehicle), with which the adaptive learning of the RBF neural network is used to estimate and approach the upper bound of the uncertainty and disturbance induced by hydrodynamics so as to avoid the difficulties of establishment and resolving of precision dynamic model. According the description and setting up of the control model, a tracking simulation was carried out and a series of tests on the yaw of ROV were performed in static pool. It is proved that this control strategy is available for the ROV.