For the problem of noncooperative spacecraft approach, flyingaround and obstacle avoidance in unmanned onorbit service, the relative kinematical and dynamic model of orbit and attitude in the noncooperative spacecraft feature part body coordinate are established. The trajectory planning is carried out, with the objective function of optimal time and fuel, and the constraint of dynamics and path. The continuous optimal control problem is discretized by the Gauss Pseudospectral method. Then the optimal path is obtained by solving the transformed nonlinear programming problem. The sixdegreeoffreedom coordinated position and attitude trajectory tracking error model are established, and the trajectory tracking control law is designed. The results of the closedloop simulation show the effectiveness and stability of the control law.

For the highprecision autonomous guidance of hypersonic vehicles with the non nominal reentry mission, a fully autonomous adaptive guidance method based on online trajectory planning and tracking law online calculating is studied. The reference trajectory with the multipath constraints is generated online via the quasi balanced glide condition and a higher precision planning model. And the feedback gain matrix of the linear quadratic regulator is online calculated when tracking the reference trajectory, obtaining a high precision adaptive tracking law. Finally, Monte Carlo simulation is presented to validate the high accuracy and robustness of the proposed algorithm, with the longrange and shortrange reentry missions.

A parallel extended Kalman filter (PEKF) algorithm for autonomous navigation of satellite constellation with multiple earth satellites and one lunar satellite is proposed. By solving the measurement selection problem with noise statistic uncertainty, the appropriate measurement is selected to reduce the exterior disturbance. By solving the measurement scheduling problem in the case of noise statistics uncertainty, the appropriate measurement is selected to reduce the impact of interferenceIn order to adaptively select appropriate measurements, the PEKF based on multiple subsets of measurements from different sources is proposed. Each extended Kalman filter (EKF) is used to process different measurement subsets, and the weight of the sub filters is calculated based on the residual sequence, compared with EKF and traditional multi model adaptive estimation (MMAE) algorithm, the results show that the proposed method is stable in the threeaxis position estimation error under the condition of interference, which reflects the advantage of the performance.

The sunsynchronous frozen orbit has the same light properties and orbital altitude as it passes under the same geographic location, which is helpful for load observation and calibration. For a constellation composed of the sunsynchronous frozen orbit, the inclination, height, eccentricity vector and relative argument should be maintained at the same time. An orbit inplane maintenance strategy based on trace monopulse with minimum fuel consumption is proposed. The stability and fuel consumption optimality of the strategy are proved, and a simple algorithm for local time maintenance of sunsynchronous orbit is given. Simulation results show that the constellation members can capture and track the target orbit after using this strategy. The inplane maintenance strategy can also be used for formation control of LEO satellites with minimum fuel consumption.

The application of lowcost micronano satellite clusters to perform space missions has drawn considerable attention in recent years. In this paper, the artificial potential field method is combined with the speed feedback control law based on the dynamics equations of relative motion for the loose formation control task of the satellite clusters. An attractive potential field is established between each satellite and the target. The intersatellite collision avoidance potential field is established based on position information. The obstacle avoidance potential field is established near space obstacles. In addition, the speed feedback control law is adopted to comprehensively realize the loose formation control of satellite clusters. Some simulations are presented to verify the effectiveness of the proposed control method.

Aiming at the problems of slow convergence speed and poor anti disturbance ability when classical PID control is applied to position and attitude control of underwater vehicle, a seriesparallel PID control method (SPPID) based on singular perturbation method is proposed. Firstly, the time scale decomposition method is used to obtain the fast and slow subsystem model of the underwater vehicle, and the SPPID is designed according to the singular perturbation method. Secondly, based on the selfdesigned underwater vehicle, the least square method is used to measure the hydrodynamic parameters. Finally, simulation and experiment show that SPPID has faster convergence speed and higher robustness.

Aiming at the control problem of space robot with load capture, a sliding mode variable structure control method based on target mass observer is proposed, considering the joint flexibility of manipulator and the unknown mass of noncooperative target. Firstly, aiming at the problem of unknown quality of space non cooperative target, an improved least square iterative algorithm is designed to realize online identification of noncooperative target quality. Then, the dynamic model of flexible joint space robot on floating base is established by using Lagrange method and the principle of moment of momentum conservation. According to the rigid flexible coupling characteristics of the dynamic model, the system model is approximately decomposed into fast and slow varying subsystems based on singular perturbation theory. For the slow varying subsystem, the sliding mode variable structure controller is designed to ensure the robustness and dynamic characteristics of the system. For the fast changing subsystem, the feedback compensation control algorithm based on the speed difference is designed to suppress the residual vibration of the flexible joint and ensure the control accuracy. Finally, the mechanism of fast and slow subsystems is analyzed by experiments, and the effectiveness of the control method is verified.

With the development of aerospace science and technology, intelligent fault diagnosis technology is one of the key technologies to ensure safe and autonomous operation of spacecraft control system. Due to the small number of unlabeled telemetry data samples with high noise, it is difficult to diagnose fault signals accurately of the spacecraft in orbit by traditional fault diagnosis methods. A deep transfer learningbased fault diagnosis method is proposed to realize realtime fault diagnosis of spacecraft in orbit. First, onedimensional timedomain signals are converted into twodimensional image signals to realize the preprocessing of spacecraft operation dataset. Secondly, a residual networkbased deep learning fault diagnosis framework is built and pretrained via ground trained dataset and onorbit operation data of other spacecraft. Then, in order to realize realtime fault diagnosis of current spacecraft in orbit, parameters of the fault diagnosis model are readjusted to adapt the model to the current spacecraft fault diagnosis. Simulation results show that the proposed deep transfer learningbased fault diagnosis method can diagnose spacecraft fault signal quickly and accurately.

Aiming at the problem that the speed loop of the permanent magnet synchronous motor vector control system adopts the PI regulator cannot adapt to the change of the motor parameters well, the control effect is not ideal. On the basis of the current predictive control (CPC) heightening the antiinterference ability and improving robustness, a sliding mode control method is proposed. The control method combining sliding mode and current prediction improves the dynamic performance of the permanent magnet synchronous motor vector control system, realizes fast speed regulation response when the load is disturbed and no overshoot control; and improves the system’s tolerance to internal resistance changes. The model of the permanent magnet synchronous motor vector control system is built in Simulink. The simulation results show that under the control of the method in this paper, the response speed and stability of the system are improved, and the time taken for the speed response from starting to stable is reduced by 60%.

The adhesion mechanism of gecko foot is of great significance to the development of space robot foot. In this paper, based on the adhesion model between a single seta and smooth surface, the distribution model of satellite rough surface, the contact probability and adhesion model between the seta and satellite rough surface, the adsorption capacity of gecko bristle array to satellite surface is calculated and analyzed. It is found that under certain conditions, simulated gecko bristle array can adhere to the satellite surface in the weightlessness environment of space. It is of significance for the research and development of the climbing robot and pipeline robot in space.