On the fuzzy-adaptive command filtered backstepping control of an underactuated autonomous underwater vehicle in the three-dimensional space

This paper studies the three-dimensional path following control problem for an underactuated autonomous underwater vehicle in the presence of parameter uncertainties and external disturbances. Firstly, an appropriate model for the error dynamics was established to solve the path following problem in a moving Serret-Frenet frame. Secondly, an adaptive robust control scheme is proposed through fuzzy logic theory, command filtered backstepping method and an adaptation mechanism. Finally, a suitable Lyapunov candidate function is utilized to verify the stability of the overall control system and demonstrate uniform ultimate boundedness of path following errors. Following novelties are highlighted in this study: (i) The fuzzy method is adopted to solve the problems of model uncertainties, which makes the controller more practical; (ii) to calculate the virtual control derivative, a second-order filter is designed. This reduces the computational effort of the standard backstepping technique. Moreover, the effect of high frequency measurement noise is considerably attenuated via an appropriate filter to attain a more robust control system. (iii) To attain a desired approximation accuracy between the virtual control and the filtered signals, a compensation loop containing the filtered error is established. (iv) An anti-windup design is proposed to solve the problem of integral saturation in control input signals. Finally, comparative simulations are performed to ensure that the presented control scheme has excellent following accuracy and good robustness under multiple uncertainties and external disturbances.