Motion Control Algorithm Design Of SCARA Based On TwinCAT

In recent years, labor costs increase continually. Industrial field such as producing food and medicament have brought a large number of robots to perform sorting, transporting, etc. At the same time, the motional speed and positional accuracy of industrial robots are required. SCARA is widely used in the field, becase SCARA has flexible horizontal, rigid vertical, large work space, easily control, and high cost performance. Robot dynamic coupling effect is obvious when running at high speed, what’s more, load changes during handling process, the traditional PID control algorithm using forced decoupling policy, and the control gain is fixed, so it is difficult to achieve the desired control performance. According to the characteristics of each axis coupling design control algorithms, the robot can improve the accuracy of high-speed operation, in which case the controller needs a strong operational performance and openness. In this paper, SCARA as the research object, build Twin CAT-based motion controller, which has fast computing speed, high realtime and good openness, design model-based algorithm and intelligent algorithm to improve the speed and accuracy of SCARA.Firstly, the positive kinematics and inverse kinematics of SCARA is analyzed and its correctness is verified by MATLAB simulation. Carry out the robot Jacobian matrix, Hessian matrix solving and analysis singularity, then utilize the end of the robot manipulator in Cartesian space trajectory planning to give joint space angle, velocity and angular acceleration correspondingly. Solve SCARA dynamics model by Lagrange method.Secondly, to design the model-based control algorithms need access to accurate and complete dynamic model by dynamic parameters identification. But there is random error during the identification, so propose a two-step identification method to determine some parameters first, which helps to select the remaining parameters of identify results. Use Lu Gre dynamic friction model as joint friction model of the robot, identify the coefficient of friction model, and obtain the relationship between friction and joint velocity. Further give the linear dynamic model of the robot with measurement matrix and a set of minimum inertial parameters, and then design incentive track and experiment to obtain the robot inertial parameters identification.Thirdly, in order to improve the robot trajectory accuracy at high speed, designed model-based trajectory tracking control algorithm. However, load changes during handling process would lead to dynamic model is inaccurate. Trajectory tracking accuracy of computed torque control method is not enough while robot modeling is inaccurate. So improved scheme is proposed using fuzzy compensation, further as the goal to response rapidly with a load disturbance, propose a fuzzy self-adjusting compensation scheme.Finally, set up the experimental platform, write control algorithm, verify the feasibility of the algorithm, test robot tracking accuracy while running at high speed. Experiments show that the same trajectory, computed torque control method required maximum torque is smaller than PID algorithm, so the long running will save energy and help to further improve the speediness of the robot. Computed torque control algorithm plus fuzzy compensation and variable gain fuzzy compensation control improve tracking accuracy. The speed and accuracy of SCARA robot is improved ultimately.