Formation Control Of Multi-satellite Simulator Based On Nonlinear Prediction Strategy

As humans successfully stepped into space,the application of satellites has become more and more extensive,and satellite formations composed of multiple satellites are easier to complete complex space missions than a single satellite.In the system that composes the satellite formation,it is necessary to consider the structural characteristics of the system,that is,the controller unit,sensor components,and actuators are distributed independently in space,so it is necessary to realize the transfer of control information and status information between each unit.,You need to use a wireless network to achieve.This paper proposes a cost function for networked nonlinear multi-agent control systems.The cost function not only considers the coordination performance between agents,but also considers the performance of each individual.In this paper,the air bearing table is used as the controlled agent,a suitable formation method is designed,and a data-driven predictive control scheme is proposed to control the formation system,actively considers communication constraints,minimizes the cost function,and implements the control algorithm.A simulation experiment was carried out to verify its control effect,and a full physical simulation experiment was carried out on the air bearing platform on the NCSLab experimental platform,which verified the effectiveness of the method.First,the structure and operating characteristics of the satellite simulator are introduced,and some parameters of the satellite simulator are directly or indirectly measured through experiments to obtain physical data.With the help of the knowledge of basic mechanics and rigid body dynamics,the inertial coordinate system and the body coordinate system of the satellite simulator were established on the basis of the marble platform used in the experiment.The state of motion on the experimental platform was described,and the formation theory was introduced.The basic knowledge related to predictive control theory puts forward the concept of cost function used in data-driven predictive control.Secondly,according to the physical characteristics of the satellite simulator,the kinematics and dynamics equations of the satellite simulator are established based on the two motion modes of the mass point translation and the rigid body rotation,and the appropriate physical quantities are selected as the state variables to establish the satellite simulator’s The state space model is discretized,the nonlinear model of the satellite simulator is established,and the parameter design interface is designed.According to the characteristics of the controller and formation needs,combined with the advantages of the pilot follower method and the virtual structure method,the virtual trajectory method is designed,and then the controller based on the data-driven predictive control is designed,and based on the input of the nonlinear model of the satellite simulator The output quantity determines the input and output signals of the controller module,verifies the stability of the controller,and designs the auxiliary controller module according to the requirements.Then a data-driven predictive control controller is used to control a single satellite simulator,and the trajectory tracking simulation experiment of linear trajectory and sinusoidal trajectory is completed,and the best controller parameters for tracking different trajectories are determined through experiments.After the formation method is used to determine the formation,the satellite simulator is formed to control and track the target trajectory.The linear trajectory and the sinusoidal trajectory are selected as the target trajectory,and the optimal controller parameters in the formation control are determined,and active compensation is considered.The possible influence of data loss caused by network delay and data loss verifies that the controller has a better control effect.Finally,through the NCSLab networked laboratory and the air-bearing platform full physical simulation experiment platform,the satellite simulator is used to perform full physical experiments on the satellite simulator using data-driven predictive control,so that the satellite simulator can track the target linear trajectory steadily and quickly,and has a better performance.Good anti-interference ability.