Research On Trajectory Planning And Servo Control For 6-DOF Stewart Platform

6-DOF Stewart platform has been widely applied to the motion simulation equipment,with traits such as high speed,large load and high accuracy.The requirement of accuracy and dynamic performance has become higher with the development of motion simulation.In order to accomplish the close loop control in work space,position and pose of the moving platform should be detected real-timely.In the existing control system,techniques such as laser tracker and grating ruler increased the cost of control system greatly.Then real-time forward kinematic problem should be studied.What’s more,unreasonable trajectory for end of the platform in practical application can cause negative influences such as affecting accuracy of motion and even damage to the mechanical structure,which reduces of the service life of the platform.Therefore servo control techniques include real-time solution of direct kinematic problem and optimal trajectory become quite important.A 6-DOF Stewart platform driven by ac servo motor was regarded as the research object,problem of real-time solution for direct kinematic problem,jerk-optimal trajectory planning in work space and joint space and servo control technique has been researched in this thesis Content of this paper is studied from the following aspects:Function and performance index of the platform have been analyzed and scheme of the control system were accomplished.IPC and PCI cards with digital pulse function,digital I/O function and encoder acquisition function were applied in this system.Interface signal process circuit were designed,which realized the magnetic segregation function and electric level match function.Program frame of servo control software were designed based on hierarchical programming principle,function model were accomplished and technical difficulties were summarized.Theory of the kinematic problem of Stewart platform were analyzed.Newton-Raphson iterative algorithm with alterable search origin used for direct kinematic problem were raised.Calculation program was developed in C++ and simulation test were finished.The results showed that solution of the direct kinematic problem can be finished in microsecond level,which laid foundation for the real-time closed-loop control of the platform.In order to improve the influence on mechanical structure caused by jerk and vibration of the platform,realize start-stop without vibration and smooth motion,cubic splines were applied to the trajectory planning in work space and joint space.In order to make the velocity and acceleration at starting point and end point controllable,a special interpolation that adds two virtual point was used.And SQP-F algorithm was used to solve the jerk-optimal trajectory planning problem.Test program was accomplished in C++ and the test results verified the validity of the algorithm.Control system of the platform is accomplished based on researches above.Software of the system is developed with C++ in Windows XP and timing function of the software is tested.Contrast experiment on jerk optimal trajectory planning and cubic spline trajectory planning is accomplished.Command signal contains translational motion signal,rotational motion signal,combination of translational and rotational motion.The result proves that jerk optimal trajectory planning reduces error of the motion control.Sine signal was used to test performance of the system and circular motion was simulated at last.