Research On The Key Technology For The Open Control Platform Of Parallel Robots

Industrial robots are essential equipment for the automated manufacturingproduction line. It can instead of people to complete repeated long-time operations, orexecute tasks in dangerous harsh operating environment, which will greatly improvedproduction efficiency. For the purpose of making the robots having the power tocomplete more complex tasks and enhance robot intelligence, many sensors, such astactile sensors, force sensors or vision sensors, and many advanced control algorithms,such as fuzzy control, robust control, need to be integrated into the industrial robot, formaking it has the ability to identify or further adaptive, self-learning ability. But thenormal commercial robot platforms are not openness, the transformation of the platform,need the assistance of developers and equipment specified by the developer. And thistransformation is not only time-consuming, but also need great effort. Sometimes thistransformation is even impossible.In this thesis, based on the analysis of the robot control platform, in order toachieve openness and real-time requirements of the platform, providing a basis foradvanced control algorithms and multi-sensor fusion, first, the structure of combinationof industrial computers and embedded motion controllers are proposed, bothcommunicate through network. Industrial computers are used as the main controller,which perform non real-time tasks. Embedded motion controller is compatible with theIEC61131standard, which are used to completing for robot control. And according tothis structure, hardware construction and software design of the platform are completed.Then, intending to improve the operation speed and control performance of therobot, the research of the path planning and path tracking has been done. For the typicalpick-and-place operation path of parallel robots, the elliptical path with modified sinemotion profile is purposed, and the dynamics based time-optimal motion planning isderived. This motion is smooth and suitable for high speed applications to avoidvibration. By comparing robot control algorithms, in the analysis of the pros and cons ofkinematics and dynamics control, a control algorithm, computed torque controlalgorithm which has a better performance, is used. A simulation has been established toverify the effectiveness of the control algorithm. For providing a basis for dynamiccontrol algorithms and optimization algorithms, a full Delta robot dynamics model hasbeen derived. Finally, in order to verify the performance of the platform and the accuracy of theDelta robot model of this thesis, some related experiments have been done on theestablished platform. The comparison of the trajectory error and the torque of the twocontrol algorithms, two motion planning and time-optimal algorithm are presented. Theresults show that the performance of the Delta robot has been greatly improved by usingthe computed torque control algorithm and elliptical path motion planning with thetime-optimal algorithm.