Research On Dead Reckoning And Adaptive Control Of Underwater Glider

Hybrid underwater gliders(HUG)were widely used in marine environment observation,military and other fields.Combining actuation of both buoyancy and propeller,the HUGs are more maneuverable and quieter,and also have the advantages of AUV(Autonomous Underwater Vehicle)with high mobility and speed.Beyond achieving various underwater applications,the motion control of HUG is one of the essential problems in attaining underwater operational objectives.However,the uncertainty of the HUG's dynamic model and the cluttered ocean environments make a great challenge for navigation control of HUGs.To deal with uncertain external disturbance and model nonlinearities,the fuzzy sliding mode control algorithm(FSMC)based on heading compensation,the dead reckoning algorithm based on extended Kalman filtering and particle filtering(EKF-PF)and the new path tracking control strategy were proposed.Subsequently,the performance of the proposed controller was illustrated through mumerical simulations results.The main contents of this thesis are as follows:Firstly,the dynamic model of a hybrid underwater glider named as Petrel-II 200 was modeled.A new fuzzy sliding mode control algorithm(FSMC)based on heading compensation control algorithm was carried out to handling with steady-state error in heading control and the chattering phenomena in traditional SMC with guaranteed heading control precision.Secondly,to overcome the shortcomings of traditional dead reckoning,such as cumulative error and low navigation accuracy,an extended Kalman filtering algorithm was presented for data fusion in dead reckoning.And the estimate of the EKF was used as the empirical value of the PF which could obtain a higher accuracy in dead reckoning.Then,aiming at the path tracking control of underwater glider with internal and external disturbances,a path tracking control strategy including Integral line of sight(ILOS)algorithm,a new FSMC algorithm based on heading compensation algorithm and a new sensor filtering with an adaptive fuzzy Kalman filtering algorithm(AFKF)was proposed.The AFKF added in the feedback control was designed to resist both process noise and measurement noise,which could realize robust adaptive following along a predefined path.Furthermore,a new depth control algorithm with anti-windup active disturbance rej ection control algorithm(ADRC-AW)was put forward.Finally,the numerical simulation was carried out to verify the effectiveness of the underwater glider's dead reckoning and adaptive control algorithm.The results showed that the proposed algorithm could effectively improve the navigation control accuracy of the underwater glider under the internal model uncertainty and external environment disturbances.