Research On Formation Control Of Air-bearing Simulators Based On Distributed Network Predictive Control

For a long time,the space missions have been characterized by high cost,high risk and long development circle.So it is necessary to take sufficient and comprehensive simulation experiments to validate the effectiveness of control strategy,especially the comprehensive distributed control strategy for satellites formation control.The air bearing simulators is a typical kind of satellites simulators,which can play the role of controlled plant in the space control system and realize testing the spacecrafts control strategy on the ground.So the design and validation experiment of distributed network prediction control strategy based on the air bearing simulators are of important significance.In this paper,we will first introduce the control theory,which includes three parts: 1.the slinding mode control theory? 2.the network prediction control theory? 3.statellite formation strategy.All of these theory introduction lay a solid theoratical foundation for the following design,including the design of sliding mode controller for a single air bearing simulator? the design of distributed network predictive controller,which can compensate the network communication delay actively and realize distributed cooperation control of air bearing simulations,on the basis of the former sliding mode controller.The third chapter mainly introduces the derivation of the distributed predictive control law used in this paper.At the beginning of this chapter,the mathematical model of the air bearing simulators are built,which will play the role of controlled plant in the mathematical simulation? then the controller for one single air bearing simulator will be design based on the slinding mode control theory? the design of multi-step states predictor is following,which will be used to predict the output and other important values of the system.All of these prediction values will be used in the design of controller to compensate the network communication delay actively? lastly,under the guidence of the designed performance index,the cooperation term of the control law is designed to get the optimal value of the performance index function.So far,the distributed cooperation control law is got,which is improved from the control law for one single simulator.The last two chapters of the paper is about the experiment part of this research.Firstly,the experiment platform and the design of software models are introduced.Then the experiments are token place based on these hardware and software platform.The experiments include the mathematical simulation based on simulink and the semi-physical simulation experiment based on Net Con,the network control system.The mathematical simulation and semi-physical simulation include two parts respectively: 1.Using the improved sliding mode controller through the network predictive control to realize the control of one single air bearing simulator,and the simulators will track the expected signals respectively to realize the formation control? 2.Used the distributed network control strategy the realize the formation control of air bearing simulators.By comparing the results of several experiments,it is found that the proposed distributed networked predictive control law can realize formation control.Although limited by the physical conditions,the control accuracy of the physical experiment is reduced compared with the digital experiment,but the other characteristics are consistent with the digital simulation,and the control accuracy is also within the acceptable range.