Date-Driven Tracking Control via Fuzzy-State Observer for AUV under Uncertain Disturbance and Time-Delay

This paper focuses on developing a data-driven trajectory tracking control approach for autonomous underwater vehicles (AUV) under uncertain external disturbance and time-delay. A novel model-free adaptive predictive (MFAPC) based fuzzy state observer (FSO) was designed to achieve high precision. Concretely, the mathematical model of AUV motion analyzed, simplified via decoupling, thus providing basis with an explicit physical explanation controller. Second, MFAPC scheme multiple-inputs multiple-outputs (MIMO) discrete time system derived, that estimates disturbance. The controller can online estimate time-varying parameter pseudo-Jacobian matrix (PJM) establish equivalent space data-model system. Third, Takagi–Sugeno (T–S) combine first time, which used decline error generated by In addition, stability proposed analyzed. Finally, two scenarios were verify effectiveness robustness FMFAPC scheme, simulations are implemented using realistic parameters T-SEA AUV.