Font Size: a A A

Modeling And Control Of Stewart Platform

Posted on:2021-04-19Degree:MasterType:Thesis
Country:ChinaCandidate:Y L HeFull Text:PDF
GTID:2428330605462321Subject:Engineering
Abstract/Summary:PDF Full Text Request
The Stewart platform consists of an upper platform(moving platform),a lower platform(fixed platform),and a connecting rod.The upper and lower platforms are connected by a hinge to achieve 6 degrees of freedom movement.As a parallel machine structure,due to its advantages of high structural rigidity,high position accuracy and high load rate,it is widely used in motion simulators,optics,precision positioning,clamping systems in electronic component production,and precision machine tools.Areas of tool control and drilling and milling.This paper focuses on the Stewart platform driven by a DC servo motor as the power element.The dynamic model of the Stewart platform was established according to the Lagrangian method,and a controller based on the inverse dynamic model was designed to realize the control of the Stewart platform based on the establishment of the dynamic inverse model.For the motion Jacobian matrix between Stewart platform's task space and joint space,it changes with the change of platform posture.The outriggers obtained through the inverse solution of posture generally perform different speed-changing motions,and it is difficult for general position feedback control methods.The dynamic synchronization of the movement trajectories of each hydraulic cylinder is ensured,so that the speed of each leg is not synchronized,which causes the problem of coupling and additional interference forces that deviate from their ideal load states.The Lagrangian method was used to establish the dynamic model of the Stewart platform,and based on feedback suppression of variable load interference,a complete inverse dynamic model of the system was established,and the ideal joint driving force trajectory of the platform was calculated based on this,Add local feedback and integral separation PID control to design the controller,so as to achieve compensation for interference force.Finally,the effectiveness of two methods of fixed-point motion and continuous motion is verified by simulation.By establishing the Elman neural network model,the solution to the positive pose of the Stewart platform is solved.Aiming at the problem that the forward solution of the Stewart platform is difficult to solve,when the Newton-based iterative method is used,if the initial value is not selected properly,it may not be able to solve a correct posture.When using the numerical method,its slow calculation speed cannot solve the pose in time.When the BP neural network algorithm is used,the result is slow,and it is very easy to fall into calculation errors,which results in not the minimum value but the local minimum value,which also means that the solution process of this method has a high probability of failure.In this case,this paper proposes to use Elman neural network to perform positive pose solution on the Stewart platform.The reason for choosing this method is that it has good dynamic characteristics,fast approximation speed,and high accuracy.Accelerate the system response speed,can quickly and accurately propose solutions to the Stewart platform.Based on the analysis of the characteristics of the Stewart platform,an H? controller is designed to control the Stewart platform,which reduces the error and improves the motion accuracy of the Stewart platform.Aiming at the multi-input and multi-output system of Stewart platform and the strong coupling between the parameters,the system is not easy to control,and it is easy to generate oscillations.Therefore,H? controller is designed to control Stewart.The H? controller reduces the error between the motion of the Stewart platform and the expected value by successive approximation of the expected value,while avoiding system oscillation and ensuring the stability of the system.
Keywords/Search Tags:Stewart platform, forward pose solution, Elman neural network, inverse model, H_? control
PDF Full Text Request
Related items