Research On Adaptive Control Based Robot Trajectory Tracking Algorithms

Today,industrial robots are widely used in the field of automation,which can not only reduce labor costs,but also increase production efficiency.Robot trajectory tracking is an important application of industrial robots.Controller is the core part of industrial robots.Therefore,designing high-performance controllers have great research significance for improving robot trajectory tracking accuracy.The trajectory tracking algorithms based on adaptive control have high precision and high robustness.For this reason,this paper aims at improving the trajectory tracking accuracy of multi-degree-of-freedom industrial robot system and exploring the application of adaptive control algorithms in robotic trajectory tracking control problems.First,this paper analyzes the physical and spatial characteristics of industrial robots and then establishes their kinematic and dynamic models.Based on the multi-joint serial industrial robot,the kinematic model is established by the modified D-H parameter method,which expresses the position and attitude of the robot.Then,the Lagrange equation is used to establish the dynamic model which establishes the model basis for the robot control.Next,several model-based adaptive trajectory tracking control algorithms are implemented based on the dynamic model of the robot.It includes PD control and its derived adaptive parameter PD control algorithms,etc.Specially,this paper explores the application of dynamic surface control method in robot control,and based on the Back-stepping method,a dynamic surface control method is realized to modify the control input.The simulation analysis shows that the proposed method can deal with the nonlinear robot system with parameter uncertainty,and it can reduce the influence of strong coupling of the robot on the trajectory tracking accuracy.In addition,model uncertainties and external disturbances in robotic systems often have unintended effects on the tracking performance.To this end,next task is to focus on overcoming the disturbance of the robot control system.The introduction of the disturbance observation mechanism can speed up the convergence of the error and improve the trajectory accuracy.Therefore,this paper establishes a control system based on nonlinear disturbance observer,which equalizes the uncertainty factor into the form of torque,so as to adaptively compensate at the control input to achieve the suppression of unknown dynamic interference.Then,the application of fuzzy logic system in robot trajectory tracking is studied.Fuzzy control is a kind of algorithm based on approximate model which does not require robotic model information.Considering that the parameters of industrial robots are difficult to determine,therefore combined with the dynamic surface control method,an adaptive fuzzy dynamic surface control method is proposed.A fuzzy system is used to approximate the physical model of the robot,and the adaptive fuzzy dynamic surface control method is used to reduce the inaccuracy of the model parameters.Compared with the model-based control strategies and the control strategies of partial parameter approximation,the proposed method approximates the model as a whole and minimizes the influence of uncertainty.By adding a combination of fuzzy dynamic surface controller and Lyapunov type adaptive law,the algorithm compensates for the unknown time lag and can converge the trajectory tracking error to an arbitrarily small area.The designed controller ensures that all states and signals of the closed-loop system are bounded and can track the desired trajectory with high accuracy.Finally,according to the kinematics and dynamics of the ER10-C60 industrial robot,all the control schemes are simulated and analyzed by Matlab and experiments.A servodriven robot experimental platform based on EtherCAT communication is established in order to send control commands from PC.Trajectory tracking experiments are performed on the ER10-C60 robot.The position,velocity and torque data of each joint are collected and analyzed,and the tracking results of each algorithm on the robot platform is compared and analyzed.