Research On Folk-carrying-pole Recovery Of UUV Based On Model Predictive Control

In recent years,the scope of marine people explore is becoming more and more widely in pace with the development of the marine industry,as a result of which Unmanned Underwater Vehicle,known as UUV,is being widely concerned.However after the UUV completes the job instruction,it is needed to recycle the UUV to complete power supplying and data downloading.UUV recovery is a very important step in the use of UUV and its independent recovery is becoming increasingly important on the basis of security considerations thus making the controlling requirements for the recovery process more important.Consequently this paper tries to design a type of controller in the recovery process based on a new mode of UUV autonomous recovery——fork-carrying-pole recovery mode.Focusing on space motion in broad waters,vertical motion in the near wall environment and horizontal motion under current disturbance,this paper launches research on UUV's returning and docking phase by utilizing model predictive control algorithm,known as MPC.Specific research contents are as follows:Firstly,this paper briefly describes the specific process of fork-carrying-pole recovery of underactuated UUV and the process is divided into returning and docking phase.Then the six-degree-of-freedom-in-space model of UUV is established and the influence of the characteristics of the model on the design of the controller is analyzed.What's more the basic principles and characteristics of MPC is expounded and the basic knowledge of feedback linearization is introduced,thus paving the way for the following controller design.Secondly,for the returning phase of fork-carrying-pole recovery,the underactuated UUV is spatially moving in a wide area of water.By simplifying the six-degree-of-freedom-in-space model of UUV,this paper obtained the prediction model of UUV system based on state space model.Aiming at the nonlinear system,a kind of continuous time generalized model predictive control algorithm,that is,analytical model predictive control algorithm is used to process the prediction model.What's more,this paper then used the Lie derivative and feedback linearization to solve the performance index,thus obtaining the optimal control law,which is going to be the controller for UUV three-dimensional path tracking.The stability of the closed-loop system is proved by Lyapunov theory and the simulation is designed to verify the effectiveness of three-dimensional path tracking controller designed in this paper.Thirdly,for the docking phase of fork-carrying-pole recovery,the underactuated UUV ismoving in the near wall environment.The docking phase of fork-carrying-pole recovery can be divided into vertical plane and horizontal plane acquisition according to its stress condition.The docking phase is divided into vertical locating and horizontal capturing according to the underactuated UUV's stress condition.Because of the change of UUV stress sustained by vertical plane,this paper modified the UUV vertical plane motion model,the UUV vertical surface correction model is obtained.By transforming the modified model into the prediction model,this paper designed a vertical path tracking controller by using the MPC.The simulation analyses of controller are designed to verify the tracking effect of the controller by Matlab.Finally,the underactuated UUV can be disturbed by ocean currents when it is moving in the horizontal plane because of the near wall environment where the docking phase of fork-carrying-pole recovery is.As to the ocean current interference,a horizontal motion model based on the ocean current interference is established.In order to simplify the controller design,the horizontal motion model is converted into error model by introducing Serret-Frenet frame.Then the SIMO system is transformed into a SISO system by redefining output.MPC is used to design the horizontal path following controller according to the prediction model.The stability of the two subsystems respectively is proved,and the stability of the closed-loop system according to the principle of equivalence is obtained.A simulation experiment is designed to prove the effectiveness of the controller.