Motion Planning And Dynamic Control Of Delta Parallel Robot

Delta robot is a kind of parallel robot, which can mainly realize packaging, sorting and lightweight handling process. It has been widely used in electronics, food, pharmaceutical, medical and other industries. In high speed and high precision case, the dynamic characteristics of robot becomes obvious, the traditional control method based on kinematics has shortcomings of low control accuracy and needing larger control energy, at the same time, Delta robot is a strong coupling and nonlinear system, all these bring certain challenges. Robot’s parameters such as links’center of mass and quality, moment of inertia, have measurement error, and structure parameters may be partially unknown or there exist deviation, so the dynamic model built is usually just the estimate of the actual model. It is of great importance to study the impact of control performance of the dynamic model with deviation. PID controller has been the most widely used controller because of its simple principle, good robustness and high reliability. Methods of PID parameter tuning are usually based on experience or intelligent algorithm, the former is cumbersome and requires certain experience, the latter is complex and needing large amount of calculation, so looking for a simple and universal method of PID parameter self-tuning is particularly important. Aimed at solving all these problems, kinematics, dynamics, motion planning and decoupling control based on dynamics are studied, and stochastic particle method for PID parameter self-tuning is proposed.Firstly, we analyze the kinematics and dynamics of Delta parallel robot. Robot’s forward kinematics is deduced and verified by experiment; a simple and efficient method of drawing robot’s reachable workspace is proposed; the inverse kinematics of Delta is established and verified by forward kinematics; the Jacobian matrix, inverse velocity and acceleration are also deduced; Delta’s singular configuration is analyzed; combining with the characteristics of the structure of the robot, we propose a method to solve the problem of the distribution of driven arms’ quality, and the inverse dynamics is greatly simplified.Secondly, we analyze Delta’s motion planning——path and trajectory planning. In order to determine the shortest time of trajectory, a method to solve Delta robot’s velocity and acceleration limit along arbitrary path in reachable workspace is put forward; a novel path and trajectory planning method of Delta robot is proposed and the theoretical shortest time of trajectory planning is deduced.Thirdly, the torque decoupling control method of Delta robot in joint space and task space is studied. The torque decoupling control method of Delta robot in joint space and task space is deduced and it is proved that the method is global asymptotic stable in the ideal and non-ideal estimation in terms of the inertial matrix, Coriolis and centrifugal force, gravity vector; for the analysis of control performance in case of the inertial matrix, Coriolis and centrifugal force, gravity vector exist deviation, a method of applying stochastic disturbance is proposed, combined with the simulation experiment, we compare the different disturbance factor’s influence on control performance.Fourthly, the optimization index of PID parameter self-tuning is constructed, and the basic stochastic particle method and improved stochastic particle method of PID parameter self-tuning is put forward. We define stochastic particle’s direct product space and optimal iteration rule; in order to solve the problem of particle escaping, a "mirror" model is put forward; to improve the searching precision and local searching ability of stochastic particle, the improved stochastic particle method is proposed. Simulation experiments show that the basic stochastic particle method has better tracking performance than without optimization, and the improved stochastic particle method is superior to the basic stochastic particle method.Finally, using the visual interface of MATLAB software, the establishment of Delta robot’s simulation interface is finished. It provides a visual simulation platform for further research.