The Research Of Air-Bearing Simulator’s Motion Control Based On Model Predictive Control

The air-bearing spacecraft simulator test bed(air-bearing testbed,ABT)is one of the ground-based simulation devices,which allows for almost frictionless motion on a flat surface with the use of planner air-bearings.With the advantages of lower financial cost,longer time of experiments lasted and smaller disturbances,in recent decades it has founded extensive applications and been widely used in various fields of space mission research,such as control of flexible manipulator,manipulator hardware testing,free floating satellite-manipulator systems,satellite formation flying and proximity operations,rendezvous and docking(capture),tests of landing gears for low-g bodies and so on.As the basis of these studies,the motion control of air-bearing testbed should be studied first.This thesis provides an in-detail description of designing and building an air-bearing testbed motion control system(ATMCS),and attempts to use a model predictive control(MPC)controller.Initially,the basic components and key parameters of air-bearing testbed are introduced.According to the relevant principles of theoretical mechanics,the inertial coordinate system(ICS)and the body-fixed coordinate system(BCS)are established.With the analysis of air-bearing testbed dynamics,the dynamic model of the system is deduced in the ICS and the BCS respectively.In addition,some knowledge of MPC is reviewed for later controller design.Subsequently,the ATMCS is established.The software architecture of the onboard computer is designed and some basic modules are designed and implemented.Firstly,a nonlinear tracking differentiator is used as the observer,whose function is verified through simulation and experiment respectively.Secondly,the thrust transformation and distribution algorithm is designed and implemented according to the mounted position of the nozzles.Thirdly,the signal modulation is designed,for the PWM modulation parts,a classic PWM duty calculation method is given,for the PWPF modulation parts,reasonable parameters are selected and simulation verification is carried out.Besides,the controllers are designed and their effectiveness is verified.Firstly,based on the analysis of the on-off characteristics caused by the solenoid valve,the PID controller is designed,which is served as a contrast to later MPC controller design.Then,the MPC controller is attempted to design.The step signal,ramp signal and sine signal are chosen for the numerical simulations and experimental verifications.The results demonstrate that the PID controller is qualified for the experiment,and the MPC controller,limited to hardware and computation capacity of the onboard computer,is also available for motion control.Finally,some typical movements including free-flying and fixed-point floating,rectangular trajectory tracking and circular trajectory tracking are abstracted from the major application fields of ABT.The numerical simulations and experimental verifications are carried out respectively,with the application of two controllers.The results show that these controllers have achieved the control goal.Because the performance of the MPC controller is limited.The control effect of the PID controller is better than the MPC controller,whose performance is limited.But the MPC controller shows some advantages such as the more stable control process,the better anti-disturb capability,the control variables stayed within the output range of the actuator(processing the constraints actively and explicitly),a less time of the actuator dropped into saturated and a small amount of fuel costs based on less control actions.