Research On Trajectory Tracking Control Of Dobot Manipulator Based On Sliding Mode Variable Structure

With the development of science and technology,the manipulator has been widely used and developed in the fields of medical,aviation,industrial manufacturing,etc.,and its trajectory tracking performance has also been raised with higher requirements.However,the manipulator is a multivariable,strongly coupled,time-varying,nonlinear system.It is difficult to establish accurate dynamic models,which will greatly hinder the realization of stable,fast and accurate trajectory tracking control.Therefore,the research on the trajectory tracking control method of the manipulator has a high theoretical and practical value.In this paper,Dobot serial manipulator is taken as the research object,kinematics model and dynamics model of the manipulator are established,trajectory planning is studied,and then the corresponding sliding mode variable structure control method is designed to realize the trajectory tracking control of the manipulator.The specific work is as follows:(1)The kinematics and dynamics of Dobot manipulator are analyzed.Firstly,the forward kinematics model of Dobot manipulator is established by D-H modeling method,and the inverse kinematics of manipulator is solved by inverse transformation method.The correctness of the forward and inverse kinematics analysis is verified by numerical examples.Then,the Lagrangian dynamics modeling method is adopted to establish a simplified dynamics model of Dobot manipulator.(2)The trajectory planning method of the manipulator in joint space and Cartesian space is studied.Firstly,the cubic polynomial and the fifth-order polynomial trajectory planning method of joint space are introduced,and the simulation comparison shows that the fifth-order polynomial can obtain a more stable and smooth trajectory curve.Then,the trajectory planning method of linear interpolation algorithm and circular interpolation algorithm in Cartesian space is studied,and its feasibility and effectiveness are verified by simulation.(3)Aiming at the modeling error and external disturbance of the Dobot manipulator dynamics model,and the upper bound is known,a sliding mode variable structure control algorithm based on the improved reaching law is proposed.Firstly,according to the chattering problem of the sliding mode variable structure exponential reaching law,an improved exponential reaching law is designed.Then the nominal model and the uncertain part of the manipulator are controlled by the calculated torque method and the sliding mode variable structure control with improved reaching law,the upper bound value of the known uncertain part is used to compensate the modeling error and the external interference.Finally,the stability of the system is proved by the Lyapunov function.The simulation results show that the control algorithm has a good trajectory tracking effect,and the chattering suppression effect and dynamic performance are significantly improved.(4)Aiming at the large external disturbance and modeling error in the dynamic model of Dobot manipulator,the neural network adaptive sliding mode variable structure control algorithm based on disturbance observer is designed in the absence of upper bound prior knowledge.The method uses nonlinear disturbance observer to observe and compensate external disturbances,and uses RBF neural network to adaptively track system modeling errors,and determines the neural network weight adaptive law through Lyapunov function.The sliding mode controller is used as the main controller to further enhance the robustness of the system.The simulation results show that the control algorithm has better anti-interference ability and control precision,and can realize fast,stable and accurate tracking of the desired trajectory of the planning.