Research On Optimization Of Parameters And Motion Control Of DELTA Parallel Robot

With the development of Chinese economy, people’s living standards, labor costs,product safety and accuracy requirements are also rising. Highly flexible equipment,which has high speed, high accuracy and high load capacity, becomes an integral part ofautomated production line. Against the development of the automated production linesin food, pharmaceutical and consumer goods industries, in this thesis, a DELTA parallelis developed, which will work instead of man to meet the light-load, high-speed,high-precision requirements of pipeline operations.Firstly, the DELTA parallel robot mechanism is introduced in my research, whosestructure is simplified. Analytic geometry method is used to obtain the analyticalexpression. In order to verify the reasonableness of simplification and the correctness ofdirect and inverse solution, two joint simulation systems are established based onMATLAB and ADAMS.Then the singularity is analyzed, obtain the necessary conditions of singularity’sappearance, and put effective ways to circumvent various singularities. Against thepractical application of the robot, put mission work space forward to meet the actualneeds, analyze its analytical expression. The required structure parameter range isobtained for parameter optimization, which lay the foundation for subsequent structuralparameters optimization and trajectory planning.Then the simplified inverse dynamics model is analyzed. Optimization objective isproposed for minimizing the maximum singular value of inertia matrix in the wholework space, which is constrained by the size and speed of the parallel robot. Finally theoptimum structure parameters are obtained by MATLAB, and a suitable working spacein the work space is obtained, which is useful for the trajectory planning and motioncontrol.Finally, the trajectory planning is analyzed, which has short cycle time with noimpact. A computed torque controller is designed based on the simplified inversedynamics and a joint simulation system is established based on MATLAB and ADAMS.The results show that the midpoints have a low positional accuracy, but the end pointposition has a high accuracy. Since the robot mainly is used in pick and place, whosemidpoints of path don’t need high accuracy, so the controller satisfies the needs of itspractical application.