Predictive Trajectory Tracking Control of Autonomous Underwater Vehicles Based on Variable Fuzzy Predictor

A variable-fuzzy-predictor-based predictive control approach is presented to solve the dynamic trajectory tracking problem of an autonomous underwater vehicle (AUV) in a three-dimensional underwater environment. To adapt to the characteristics of AUV's motion such as nonlinearity and time-varying dynamics, a predictive controller framework is proposed based on a variable fuzzy predictor whose structure and parameters are both online adjusted. To achieve a precise estimation of AUV's motion, the variable multi-dimensional fuzzy predictor employs a sliding data window (SDW) as the system status observer, and uses the Delaunay triangulation partition method for model construction. The predictive control scheme takes advantages of the constraints-tolerance of predictive control, the uncertainty immunity of fuzzy logic calculation, and the adaptability of variable fuzzy model. The comparative simulation of the AUV trajectory tracking control is conducted in the scenario of underwater operation of an offshore platform, and the result demonstrates the feasibility and effectiveness of the proposed control strategy in respect of accuracy and stability.