Font Size: a A A

Research On Forward Kinematics Based On Hybrid Strategy And Servo Control Of Stewart Platform

Posted on:2020-04-02Degree:MasterType:Thesis
Country:ChinaCandidate:Z P YaoFull Text:PDF
GTID:2428330575968788Subject:Mechanical engineering
Abstract/Summary:PDF Full Text Request
The six-degree-of-freedom Stewart configuration is a classic configuration for parallel platforms.It is widely used in simulators,machine tools,vibration,space,and medical fields.These applications make full use of the parallel platform with excellent response characteristics such as high response speed,good dynamic characteristics,and small error accumulation.The parallel platform is a complex nonlinear system.There are still many problems in the fields of kinematics,dynamics,workspace and singularity,and control strategies.The premise of these problem is the forward kinematics solution.In this paper,the 6DOF Stewart configuration parallel platform is taken as the research object.Research on Forward kinematics Based on Hybrid Strategy that combination of iterative method and intelligent algorithm.The intelligent algorithm uses BP neural network.According to the difference of the six degrees of freedom of the dynamic platform to the neural network,the structure of three kinds of BP neural networks with different nodes and hidden layers is designed.The neural network performs training and output test,and finally obtains a superior neural structure.Then the optimal neural network is solved to obtain the approximate solution of the positive solution of the position,and then the approximate solution is used as the iterative initial value of the Newton-Raphson iterative method.The final result is used as the value of the positive solution of the position.The construction of the experimental environment is based on the parallel platform driven by the servo electric cylinder.The servo control scheme based on the general computer combined with the controller is selected.Then completed the overall design of the control system of the platform,including the design of the hardware platform and the development of the control software.Before the parallel platform experiment,path planning is performed on the motion of the platform.Then the boundary search method is used to analyze the working space of the parallel platform.The three-dimensional maps of the fixed attitude position working space and the fixed position attitude working space are drawn,further analyze the characteristics of the parallel platform working space.Then based on the Jacobian determinant,the three-dimensional space of the singular position and the singularity of the fixed position is drawn by the symbol solving software maple.it is proved that there is no singular configuration in the set working space.Finally,the SBG inertial navigation and hybrid strategy position positive solution are used to error calibration the parallel platform.In the actual control experiment of the parallel platform,the two motion functions of static positioning and dynamic following are realized.The positioning accuracy and motion following error are measured and analyzed to verify that the effectiveness of hybrid strategy applications and the rationality of path planning.
Keywords/Search Tags:Parallel Platform, Hybrid strategy, Forward kinematics, Path plan, Workspace, Singularity
PDF Full Text Request
Related items