Research On Coordinated Control Of Autonomous Underwater Vehicle-Manipulator Systems

This dissertation focuses on the deep study of the coordinated control of autonomous underwater vehicle-manipulator systems(UVMS),which contains two aspects:the motion planning and the trajectory tracking control in the joint space.Since the UVMS is kinematic redundancy and each subsystem has different dynamic response characteristics,a fuzzy logic technique based multi-tasks weighted gradient projection method is proposed which effectively solves the motion planning problem of the UVMS.Meanwhile,for the trajectory tracking control problem of UVMS in joint space with strong nonlinear couplings,complex parametric uncertainties and unknown external disturbance,a novel time delay estimation(TDE)based nonlinear robust control strategy is proposed and investigated which effectively solves the nonlinear robust trajectory tracking control problem in joint space of UVMS.Finally,the coordinated control problem of UVMS under large lumped uncertainties is effectively solved by combining the proposed motion planning algorithm and the nonlinear robust control method.This dissertation is divided into eight chapters,each chapter is summarized as follows:In chapter one,the history and current domestic and international situation of UVMS is briefly introduced.Moreover,detailed review and introduction are given for the coordinated control of UVMS including motion planning and trajectory tracking control in the joint space.Finally,the research significance,research difficulties and research contents are given briefly.In chapter two,the development background,overall structure design and each subsystem of the UVMS are introduced.Meanwhile,the fitted equation between generalized control input signal and output force of the propellers are given through pool experiments.In chapter three,the kinematics properties of underwater vehicle and underwater electric manipulator are briefly introduced and corresponding kinematic models are presented,respectively.Afterwards,the kinematic models of UVMS in joint space and task space are successfully obtained combining the ones of underwater vehicle and underwater electric manipulator.Usually,it is very difficult to obtain the dynamic model of UVMS through rigorous mathematical derivation considering the high freedom of degrees(DOFs)and the complexity of hydrodynamics.Therefore,the dynamics model of UVMS is presented using the SimMechanics toolbox under MATLAB/Simulink@ and Solidworks.Finally,integrated simulation model of UVMS is given under the MATLAB/Simulink@ environment.In chapter four,a fuzzy logic technique based multi-tasks weighted gradient projection method is proposed for the motion planning problem of UVMS which is kinematic redundancy.The proposed method effectively solves the priority problem between the secondary tasks using the fuzzy logic technique.Then,the weighted gradient projection method is adopted to realize the main task,meanwhile the secondary tasks can also be well handled.Finally,the effectiveness of the proposed method is demonstrated through several comparative simulations.In chapter five,for the trajectory tracking control problem of UVMS in joint space with strong nonlinear couplings,complex parametric uncertainties and unknown external disturbance,a novel continuous TDE based nonlinear robust control strategy is proposed.The core idea of this strategy is to use the TDE technique as the baseline part of the overall control algorithm,afterwards other robust control method is adopted to adjust the control performance.Meanwhile,the core idea of TDE technique is to estimate the current lumped unknown dynamics of the plants directly using the intentionally time-delayed information of the closed-loop control system,afterwards the system dynamics is compensated using the obtained current lumped dynamics.Therefore,the TDE-based controllers usually require very few information of the system dynamics.Thus,the author firstly proposes a TDE-based PD controller,which is simple and easy to use.Afterwards,the terminal sliding mode(TSM)control method is combined with the TDE technique to further improve the control performance of the system.Stability of the closed-loop control system under lumped uncertainties is analyzed using the Lyapunov stability theory,and corresponding tracking errors are also given.To demonstrate the effectiveness of the proposed control methods,several comparative simulations and pool experiments have been conducted using the proposed two TDE-based control methods and the traditional PD controller.Corresponding results show that both proposed methods can effectively ensure more satisfactory control performance and better robustness compared with the traditional PD controller,meanwhile very few information of the system dynamics are required.Furthermore,the proposed TDE-based TSM control method can obtain better control performance and stronger robustness compared with the proposed TDE-based PD controller.In chapter six,to further improve the control performance and ease the influence from the measurement noise for the trajectory tracking control problem of UVMS in joint space with large lumped uncertainties,a novel discrete TDE(DTDE)technique based nonlinear robust control method is proposed which is obtained based on the proposed continuous TDE based nonlinear robust control strategy.Usually,UVMS will not be equipped with the acceleration sensors.Thus numerical differentiation of the position or velocity signal is required to obtain the acceleration signal needed by the continuous TDE technique.Although the influence of the measurement noise can be reduced through lowering corresponding control parameters or introducing some filters,this kind of operations can still greatly limit the adjustment effect of the inner controller.Meanwhile,the numerical differentiation and extra filters will result in a complex control structure and require additional computational effort,which is not good for the practical applications of the continuous TDE-based proposed controllers.Compared with the continuous TDE-based controllers,the ones based on DTDE technique no longer require the acceleration information leading to a simpler structure and more suitable for practical applications.Stability of the closed-loop control system is analyzed using the Lyapunov stability theory,which shows that the trajectory tracking errors will converge to a small ball field.To verify the effectiveness of the proposed method,numerical simulation and pool experiment are conducted.Corresponding results show that the proposed control method can ensure satisfactory control performance with a relative simpler control structure compared with the continuous TDE-based controllers,which is more suitable for practical applications.In chapter seven,for the coordinated control problem of UVMS under large lumped uncertainties,a coordinated control strategy for UVMS is proposed combining the proposed motion planning method and the nonlinear robust controller effectively.The effectiveness of the proposed strategy is demonstrated through simulation.In chapter eight,the main research work,results and innovations of this dissertation are summarized and analyzed.Further work and the research orientation of the coordinated control of UVMS are briefly given.