Error Compensation And Modal Decoupling Control Strategy Of Parallel Platform

As the continuous progress and development of China's industry,the position of robots in the field of industrial automation is also constantly improving.Mobile industrial assembly robots can replace humans in complex and high-risk environments,effectively improved the quality of products.Typical assembly robots are mostly of serial robots,which have the low load capacity and are only suitable for the assembly of light parts.Compared to robots with serial structure,the parallel robots are characterized by high load-bearing capacity,high positioning accuracy and high stiffness.The application of parallel robots in assembly condition can effectively solve the problems of heavy workpiece loads and unstable long-term transportation.When the robot is used for workpiece manufacturing and component assembly,the final quality of the product is greatly influenced by the robot's positioning errors.It is expensive and limited to improve the overall accuracy of the platform only by improving the machining and assembly accuracy,while the use of kinematic calibration method to compensate the error of the parallel platform is an effective and economical technical means to improve the positioning accuracy.Compared with the serial structure robots,the mathematical modeling process of parallel platform is more complex.And due to the inherent characteristics of the parallel platform,there is strong coupling and nonlinearity between the legs of the parallel platform,which will seriously affect the overall control performance of the parallel platform.In order to ensure the positioning accuracy and tracking performance of the parallel platform,combined with the working environment of assembly and docking,this paper focuses on the error compensation method and decoupling control strategy of the parallel platform.The main research work is as follows:(1)The positive and negative kinematic solution models of the parallel platform are presented,and the solution formulas of velocity and acceleration are derived.The inverse dynamic formulation of the parallel platform is established using the NewtonEuler method based on the equation of velocity information in kinematics.Then the inertial parameters in the dynamic model are identified based on computer technology.(2)In order to reduce the positioning error of the parallel platform,a kinematic calibration method of the parallel platform is proposed.According to the closed-loop characteristics of the branch chain of the parallel mechanism and the total differential theory,a pose error mapping model based on inverse kinematics is established.Then,considering the low computational stability of the ordinary least square method,the total least square method is chosen for the parameter identification of the parallel platform,and the solution principle and calculation process of the method are introduced.Finally,the flow of the whole calibration algorithm is introduced in detail,which lays a foundation for the subsequent calibration numerical simulation research.(3)A dynamic model of the servo system is proposed and the friction of the ball screw is estimated using Coulomb and viscous friction model.After that,the concept of modal analysis in the vibration field is introduced to analyze the parallel platform,and the modal decoupling matrix is derived.Then the transfer function of the parallel platform in physical space is derived.According to the outcomes of the modal analysis,a decoupling control strategy in modal space is proposed to realize the output decoupling of each control channel,and its stability is proved.(4)A parallel platform experimental system is built,and the hardware components and structural parameters of the system is introduced in detail.Then,the kinematics calibration numerical simulation of the parallel platform is carried out,and the measurement principle of the dual camera measuring instrument and the method of solving the position and attitude of the platform are introduced.According to the results of calibration simulation,the correctness of the error model and the total least squares solution method is verified,and the parallel platform kinematic calibration method proposed in this paper can effectively reduce the positioning error of the platform.Finally,the decoupling effect and tracking performance of the modal space decoupling control strategy proposed in this paper are verified by experiments.