Study On Localization Technique Of The End Of 6-dof Manipulator Towards Satellite Assembly Operation

In the assembly process of the satellite,the traditional installation method can not meet the work efficiency and the requirements of the installation accuracy.For it's accurate and fast,the manipulator must be bound to be the inevitable choice for the assembly of satellite and other spaceflight equipment.The location of the end of the manipulator is the key to its application,and it is also the focus of research in recent years.The process of positioning involves many fields,such as sensors,computers,signal processing,and so on.This paper focus on the noise in the positioning process of manipulator,and done a lot of work on nonlinear filtering algorithm and the motion model of the manipulator end.The main contents are as follows:First,the development and application of the manipulator is briefly summarized,The influence factors of the end positioning of the manipulator are analyzed in detail.It is focused on the influence of noise on the positioning accuracy.According to the parameters of the UR3 manipulator,the model of the manipulator is established by the D-H rule.Study the forward and inverse kinematics of the manipulator.The visual observation system,the transformation of coordinate system,image processing and camera calibration process are analyzed,and the parameters of camera are calibrated through preliminary experiments.The accuracy of visual location is verified by experiments.Secondly,focused on the high nonlinear of the motion system of the end of the manipulator,Monte Carlo method is studied.On the basis of comparative analysis of particle filter and Kalman filter,an improved particle filter algorithm based on particle swarm optimization is proposed,which can improve the problem of low coincidence degree between particle prior distribution and likelihood function.The simulation results show that the algorithm has better positioning accuracy than the original algorithm.Thirdly,focused on the application of different motion models,the filtering effect of the interactive multi model algorithm is analyzed under the two filters.An improved interactive multi model algorithm by using the state discrepancy of adaptive tolerance is improved,which improves the low precision problem caused by the outliers.Simulation results show that the improved algorithm has better performance.Finally,the experimental platform of the camera,the client,the control cabinet and the UR3 manipulator is used to verify the whole experiment.The experimental results show that algorithm has excellent performance and reliable positioning.It can provide a feasible way for the end positioning of the manipulator.