Parallel Platform Based On Stewart Structure And Its Application In Attitude Stabilization

Stewart platform is type of parallel manipulator that has six degree of freedom.It has the advantages of strong load capacity and high motion accuracy.It is widely used in motion simulator,parallel machine tool,aerospace and other fields,and is a research hotspot of parallel robot.In this paper,the Stewart platform is taken as the research object,the characteristics of the Stewart platform are studied,and the application of the parallel mechanism as the attitude stabilization platform in the bumpy environment is discussed.Kinematics modeling is an important part of parallel robot research.In this paper,the kinematics model of Stewart platform is discussed,including the rotation error of the electric cylinder linking the upper and lower platform.Based on the kinematics model,a complete simulation system of Stewart Platform in Simulink environment is realized,the linear system identification of the simulation model is completed,and the optimal controller is designed based on LQR principle.Than we verified correctness and effectiveness of the kinematics model and optimal controller discussed in this paper by trajectory tracking simulation experiments.Accurate calibration of kinematic parameters is of great significance to improve the control accuracy of Stewart platform.The existing calibration methods often need to use expensive instruments such as laser tracking interferometer system to measure.In this paper,an on-line calibration method of Stewart kinematic parameters based on computer vision is proposed,which uses ordinary optical camera and calibration pattern.When the transformation relationship between the camera coordinate system and the Stewart platform coordinate system is unknown,the parameters can be calibrated only by the checkerboard calibration pattern and the relative pose change of the Stewart platform itself.The experimental results show that it can effectively and quickly eliminate the initial length error of legs in our physical prototype.In this paper,a stable platform based on Stewart parallel structure is constructed with the application of anti bumping.The system uses IMU as the system sensing input and brushless motor as the actuator to complete the embedded system based on EtherCAT bus,and realize a Stewart structure attitude stable platform.The experiment is carried out by using Stewart platform motion as incentive in indoor environment and real bumpy road as incentive in outdoor environment,and the experimental test data are given.Aiming at the time delay of multi-axis servo system,the measurement method of communication time delay based on EtherCAT bus is also discussed.At the same time,the realization of trajectory planning and IMU attitude synthesis method are studied.The software compensation method of the rotary error of the electric cylinder is discussed.Finally,the speed feedforward design of the controller is discussed and the effectiveness of feedforward is verified in experiment.