Kinematics Calibration Of Stewart Platform Based On The Optimal Selection Of Configurations

After the successful launch of"Shen Zhou"airship, research for the docking of spacecrafts in the space has become more important, it's the key-point which our country needs to deal with currently. To insure the success of docking of spacecrafts in the space, a mass of simulation must be done on the ground. As the core parts of simulation system, the Stewart Platform's accuracy is the most index of the system. In order to simulate the relatively movement of aircrafts efficiency and true, the accuracy of the Stewart Platform must achieve an excellence level. This paper starts with the research of technique to improve the accuracy of Stewart Platform, come down to the content of error model, kinematics calibration and optimal selection of configurations, bring forward the theory of reducing error and increasing accuracy. This work will be a support to the popularization of calibration.This paper set up the error model of the 6DOF motion simulator. A kind of exterior calibration based on the kinematics inverse problem has been studied; all the work about the calibration has been finished. First the calibration model was built up according as the kinematics inverse model; the errors of geometrical parameters were distinguished with the"least square"algorithm and iterative algorithm. Second based on the errors model, using the software"Matlab", the process of calibration was simulated, and the simulation provided the theoretic gist for the possibility of accuracy improvement. As a important factor in the calibration, a approach of optimal selection of configurations was put forward. And the simulation of the optimal selection proved the effect. Finally the experiment of pose measurement and kinematics calibration was studied out. The result of calibration simulation indicates that the calibration model is brief; and its calculating rate is high; its solution has high accuracy. The feasibility of optimal selection by genetic algorithm is proved by the contrast simulation with two group contrary configurations.