Weinan Xie

Multiple satellite formation based on multi-agent

Formation flying of multiple satellites is an enabling technology for many future space science missions. This paper addresses the problem of formation control and coordination of multiple satellite formation flying systems. Several coordination strategies for satellite formation flying are discussed. A new coordination framework based on multi-agent with advantages of robustness, flexibility and realization is proposed which considers not only the control of relative motion but also the interactions among multiple satellites in the formation. An LQR controller is presented for relative position control of satellites formation and a two-satellite formation simulation analysis was performed to demonstrate and validate the efficacy of the scheme

Robust tracking control of space robot via neural network

This paper proposes a new robust control method for space robot by using neural network. A radial-basis-function (RBF) neural network is included to compensate for the system uncertainties. The parameters of the neural network are adapted on-line according to derived learning algorithms using Lyapunov method. Simulation results of a two-link planar space robot verify the validity of the proposed controller in the presence of uncertainties

A ground semi-physical simulator for satellite autonomous rendezvous and space robot capture

The architecture of a ground semi-physical simulator is proposed, which can test the methodologies of the satellite formation flying, autonomous rendezvous and docking, and space robot capture. There are five subsystems in the ground simulator: the console system, the translation system, the rotation system, the target system, and the vision system. The console system is the most important part in the underlying system, which generates control commands based on the control algorithm. The translation system and the rotation system simulate 6 DOF motion of the follower, and the target system simulates 6 DOF motion of the target. The vision system is included in the feedback block. Through the Ethernet, the whole closed-loop system is developed and the provided real-time network communication for the system is introduced

LMI-based l/sub 2/ -l/spl infin/ stability analysis and control of networked control systems with randomly time-varying long delays

This note mainly studies stability and control of a class of networked control systems with randomly time-varying long delays. By choosing appropriate Lyapunov-Krasovskii functional and introducing the equivalent descriptor form of the system, LMI-based sufficient conditions for stability and control are derived. Finally the effectiveness of the proposed approach is illustrated by the numeric examples

Design of real-time measurement system with Vision/IMU for close-range semi-physical rendezvous and docking simulation

Design of a real-time measurement system based on information fusion of Vision Measurement Unit (VMU) and Inertial Measurement Unit (IMU) is proposed for a semi-physical simulation system. Main contributions of the paper are the design and application of VMU-IMU integration to the real-time simulation of close-range rendezvous and docking (RVD) in lunar orbit. Structure and each component of the system are presented. The pose (position and attitude) filters utilized for information fusion within Kalman filter (KF) framework are designed, which regard Hill equations and quaternion differential equations as process models respectively and choose relative pose of two spacecraft as measure variables. Capabilities of the real-time measurement system are demonstrated by semi-physical closed-loop simulations, which show the proposed system yields more precise navigation result and more effective preservation of energy than an independent VMU.

Robust energy-to-peak filtering for uncertain systems based on sampled measurement

The problem of robust energy-to-peak filtering based on sampled measurement for uncertain systems is studied. Based on piecewise Lyapunov stability criterion, the sample interval dependent conditions for the existence of such filters are established, which are less conservative. The admissible filters can be obtained from the solution of convex optimization problems in terms of linear matrix inequalities, which can be solved via efficient interior-point algorithms. Finally, a numerical example is provided to illustrate the effectiveness of the proposed techniques.

Robust H∞ filtering for uncertain systems based on sampled measurement

The problem of robust Hinfin filtering based on sampled measurement for uncertain systems is studied. Based on piecewise Lyapunov stability criterion, the sample interval dependent conditions for the existence of such filters are established, which are less conservative. The admissible filters can be obtained from the solution of convex optimization problems in terms of linear matrix inequalities, which can be solved via efficient interior-point algorithms. Finally, a numerical example is provided to illustrate the effectiveness of the proposed techniques.

Delay-Dependent Robust Sampled-Data Stability for Markovian Jump Systems with Time-Delay

This paper considers robust stochastic stabilizability for a class of uncertain sampled-data systems with time-delay and randomly jumping parameters. The transition of the jumping parameters is governed by a finite-state Markov process. The closed-loop system is a hybrid one defined on a hybrid time space and a sample space. The delay-dependent sufficient conditions on robust stochastic stabilizability for sampled-data control systems with Markovian jumping parameters are proposed using the stochastic Lyapunov-Krasovskii stability theory. The sampled-data control problems can be constructed through a set of coupled linear matrix inequalities. Finally, the numerical example is given to demonstrate the proposed techniques.

The SCIT multi-vehicle networked control testbed

This paper describes the SCIT (Space Control and Inertial Technology Research Center of HIT) multi-vehicle control system, a platform for testing networked control methodologies of multiple vehicles coordination and formation, and image processing. The system is consisted of three subsystems: the console system, the vision system and the vehicle system. The console system mainly generates the control commands through the given command information and the state information sent by the vision system, and then sends them to the vehicles. The function of the vision system is to identify the vehicle ID and calculate the state information (including the orientation and the velocity) of the vehicles. The vehicle system is consisted of some sensors and wireless Ethernet for communication. The development of the system is introduced in this paper, and the obstacle avoidance algorithm when vehicles are running is provided