Research On Sliding Mode Control Methods For Underwater Robot Manipulators

With the reduction of onshore non-renewable resource reserves,people have gradually turned their attention to the ocean,and they have developed and researched a large number of engineering equipment to perform submarine detection,salvage,submarine pipeline inspection and repair,and other tasks.As one of the important marine development equipment,underwater manipulators have attracted widespread attention.This thesis mainly studies the application of sliding mode variable structure control in underwater manipulator systems with unknown external disturbances and system uncertainties,which suppresses the chattering phenomenon in sliding mode variable structure control and improves the trajectory tracking precision of underwater manipulators.Firstly,the background and significance of this topic are elaborated in this thesis,and the development history of underwater manipulators and the theoretical progress and application of sliding mode variable structure control are reviewed.Then the Lagrange equation is used to calculate the inertia term,gravity term,and the Euclidean and centripetal force term of the non-aqueous manipulator dynamics model.Then the interaction between the manipulator and the underwater environment is considered so that a complete dynamic model of the underwater manipulator is established.Secondly,the basic concepts and control principles of sliding mode variable structure control are introduced,and the general sliding mode control method of underwater manipulators is also given.An adaptive neural network disturbance observer and an improved double power reaching law are designed to observe and compensate for unknown external disturbances and system parameter uncertainties,and to reduce chattering problem of sliding mode control systems.Based on these techniques,a fast terminal sliding mode control method is proposed to improve the convergence speed of the system trajectory.The proposed fast terminal sliding mode control is compared with the general sliding mode control to verify the suppression of chattering and the fast convergence ability.Next,considering the input saturation constraint phenomenon of the actuator,a Gaussian error function is used to design the asymmetric limit of the control torque to ensure the smoothness of the control torque.In order to avoid chattering in sliding mode control,the hyperbolic tangent function is introduced instead of the symbolic function.At the same time,the adaptive method is utilized to estimate the upper bound of external disturbances,system uncertainties and the input saturation error to improve the robustness of the system.Further,non-singular fast terminal sliding mode control is adopted to avoid the singularity in the traditional terminal sliding mode control.Finally,considering the superior anti-chattering performance of the high-order sliding mode algorithm,an improved super-twisting algorithm based non-singular fast terminal sliding mode control is designed.A high-order super-twisting disturbance observer is established to observe unknown external disturbances and system uncertainties.Inspired by the general super-twisting algorithm,an improved supertwisting algorithm is proposed and the adaptation method is given to adjust the gain parameters and applied to the non-singular fast terminal sliding mode control.Considering the existence of a time-varying gain matrix in the controller,Lyapunov theory is used to strictly verify the stability of the system.Compared with the general super-twisting algorithm based sliding mode controller,the proposed controller eliminates the prior information for the upper bound of external disturbances and system uncertainties,and has better trajectory tracking performances and chattering suppression capabilities.