The Research On High-speed Motion Control For Delta Parallel Robot

With the introduction of "Made in china 2025",industrial robots,as one of the important members in intelligent manufacturing,have been developing rapidly in recent years.At the same time,as one of the important branch of the large family of industrial robots,Delta parallel robot is widely used in high-speed sorting automation production lines in the food,medical,electronic and other industries due to its advantages of fast speed,small inertial impact and high repetitive positioning accuracy.In this paper,Delta parallel robot is taken as the research object,and its motion control algorithm under high speed is mainly studied in order to improve the overall motion performance of the robot.The main contents of this paper are presented as follows:(1)Kinematic and dynamic modeling.According to the structural characteristics of Delta parallel robot,the solution model of direct and indirect kinematics is established and correct solution is selected.By analyzing the jacobian matrix,this paper proposes three possible singular configurations and gives the solution.For dynamics,the complete and simplified dynamic model is obtained using the principle of virtual work.On this basis,the simulation platform of kinematic and dynamic algorithm validation is established.In order to verify the correctness of the forward and inverse kinematic and the rationality of dynamic simplified model,the corresponding simulation experiment is accomplished.(2)Acceleration and deceleration planning.In this paper,an improved S-type acceleration and deceleration control algorithm with speed curve as piecewise cubic polynomial is proposed,and the overall planning of this algorithm is completed.Due to the rounding of interpolation period,there maybe exist displacement deviation in the process of interpolation,therefore,a real-time interpolation algorithm is proposed.At the same time,the correctness of the acceleration and deceleration algorithm is verified through simulation experiments.Compared with other acceleration and deceleration algorithms,the result shows that the acceleration and deceleration algorithm proposed in this paper is the best in running time.(3)High-speed transition between adjacent trajectories.For the situation that robot performs series of trajectories transition,per-specified curve transition method based on elliptic curve and acceleration curve overlapping transition method are proposed in this paper.For elliptical curve transition,the mathematical model of elliptic curve trajectory point calculation is established first.Then a speed look-ahead control algorithm is proposed to make the integral velocity curve smooth.For acceleration curve overlapping transition,an adaptive planning algorithm of acceleration curve overlapping time is proposed,and make overall planning for the transfer algorithm.Based on the above discussion,the above two kinds of transition algorithms are verified by simulation and analysis.(4)Control method.According to the theory of PID control and computed torque control,Delta parallel robot control system simulation platform based on these two control methods is established in Simulink.Then through the simulation experiment,the conclusion is drawn that the control effect of the calculated torque control based on the dynamic model is better.(5)The motion control and experiment of visual tracking.Firstly,a comprehensive calibration algorithm for Delta parallel robot,vision system and transmission device is proposed.After obtaining the pose information of the target object,a preprocessing method is used to obtain it's actual position,and through the timing speed planning to complete the grasping action.Finally,a visual tracking experiment platform of Delta parallel robot is established to verify the algorithm in this paper.