Research On Trajectory Tracking Control Of Manipulator Based On Observer

With the emergence of the "Made in China 2025" national strategy,robotic arms,as an important part of manufacturing intelligence,have received unprecedented attention.With the increasing of the task difficulty and the increasing of the task in the especially complex environment,the higher accuracy of the attitude control of the end-effector of the manipulator is required.Therefore,the trajectory planning of manipulator in complex task environment is studied in this paper,and the following results are obtained.In order to solve the problem that the manipulator can't establish a fully accurate dynamic model in practical application and the manipulator is generally subject to various unknown external disturbances in the working environment,the neural network sliding mode control method based on disturbance observer was proposed.For the observable part of the external disturbance,the disturbance observer was used to eliminate it.Because of the variable parameters of the mathematical model of the manipulator system,and the randomness of the external disturbance,the disturbance observer and the mathematical model of the manipulator were combined.By selecting the design parameters,the observer can be exponentially convergent,and the disturbance can be restrained.By using the observer to estimate the disturbance output value of the manipulator system and make difference with the actual disturbance value of the system,the estimated value can be corrected and the disturbance error can be guaranteed to zero eventually.Neural network was used to approach and compensate the uncertainties in the manipulator system in real time,and neural network was introduced into the sliding mode variable structure control.A sliding mode variable structure controller based on neural network was designed to solve the uncertainties in the manipulator system and improve the corresponding tracking performance of the system.Lyapunov method was used to analyze the stability of the closed-loop control system.The simulation results show that the proposed method can overcome the uncertainties and external disturbances in the system and the manipulator can realize the desired trajectory fast tracking with excellent performance.Due to the influence of dynamic uncertainties on the manipulator,these uncertainties are difficult to predict in advance,which has a negative impact on the positioning accuracy and repeatability of the manipulator.It is considered that in many cases it is more convenient and ideal to drive the end-effector of the manipulator to follow the desired trajectory.However,in this case,precise kinematics knowledge is needed to solve the inverse kinematics problem.Because of the inaccurate measurement of physical parameters and the interaction between the manipulator and different environments,the kinematic parameters may not be known in advance.Aiming at the task space trajectory tracking control of the manipulator under theaction of kinetic and kinematic uncertainties,the effects of kinematic uncertainties,unknown moment,uncertain moment of inertia and dynamic uncertainties caused by disturbance on the manipulator are discussed in this paper.In order to accurately estimate the kinematic and dynamic uncertainties in the system,two observers were designed to deal with the kinematic uncertainties and unknown moments respectively.Two observers in the control strategy,one was used to estimate the unknown torque and the other was used to estimate the position of the end-effector.Then the sliding mode controller was designed with the estimated information to ensure the global asymptotic convergence of the trajectory tracking error.The simulation results show that the designed sliding mode observer can ensure the manipulator to track the given expected trajectory well,and the trajectory tracking error tends to zero.