Research On Adaptive Control Method Of Manipulator Trajectory Tracking

With the in-depth implementation of "Made in China 2025",manipulators have been widely used in arc welding,spraying and assembly.Manipulators are playing an increasingly important role in the intelligent process of industrial manufacturing,aerospace,rail transit and other fields.With the improvement of production technology,the tasks of the manipulator have become more complex,requiring the manipulator to achieve sufficient tracking accuracy for a given trajectory.The manufacturing industry is transforming from large-scale production to customized and personalized production,which puts forward higher requirements for the trajectory tracking and control of the manipulator.In this context,it is very important to explore the high-precision control of the trajectory tracking of the manipulator and to ensure the adaptability of the manipulator in different environments and different tasks.In this paper,aiming at the problem of model uncertainty and unknown interference of the manipulator,an adaptive controller based on fractional-order sliding mode and variable universe fuzzy is designed.The main contexts are as follows:Aiming at the chattering problem in sliding mode control,a fractional order operator is introduced into the sliding mode surface.According to the characteristics of the neural network approximator,a neural network adaptive control method based on fractional sliding mode is constructed,which improves the control accuracy and response speed of manipulator trajectory tracking,and reduces the system jitter.The fractional sliding mode function and exponential reaching law obtain the system control law.The Radial Basis Function neural network approximates the uncertainty of the system.The adaptive law is used to overcome the approximation error of the neural network in the system compensation.The error feedback signal between the desired angle/angular speed and the actual angle/angular speed is used as the controller input,and the sum of the two outputs is the input torque value of the manipulator.The simulation results show that,compared with the sliding mode neural network controller,the manipulator angular displacement adjustment time is reduced by 1.5s and 1.1s,the position root mean square error is reduced by 0.0672 rad and 0.0553 rad,the dispersion of the joint position error is reduced,and the tracking process is more Smooth and stable.Aiming at the uncertainty of the dynamic model and parameters of the manipulator,a composite control scheme combining the nominal calculation torque controller and the variable universe fuzzy compensator is constructed,which effectively improves the trajectory tracking accuracy and system robustness.Aiming at the nominal model of the manipulator,a calculation torque controller is designed;considering the uncertainty of the model of the system,a variable universe fuzzy compensator is designed to identify and compensate it.The expansion factor is a proportional factor.The scope of the universe shrinks(expands)as the steady-state error decreases(increases),which improves the convergence speed and error accuracy of the system.The adaptive law ensures that the variable universe fuzzy system is applicable to different regions of the state space at different times.The Lyapunov stability theory proves the stability of the closed-loop system.The simulation results show that the angular displacement adjustment time of the manipulator control system based on variable universe fuzzy compensation is shortened by 1.1s and 1.2s,and the position root mean square error is reduced by 0.0176 rad and 0.0133 rad,which effectively improves the trajectory tracking performance of the system.