The new generation of China geostationary meteorological satellites, FY4, includes optical and microwave sounding satellites, will present the status of multiple satellites in orbit at the same time. Ranging, orbit determination and prediction (called orbit determination) are the supporting systems of satellite image navigation, attitude maneuver and orbit management, which are characterized by high accuracy and fast response. In view of the problems and limitations of the ground orbit determination system under development, as well as the accuracy and timeliness of the geostationary meteorological satellite orbit determination using GNSS (global navigation satellite system), simulations are carried out, and the parameters such as the number of satellites that BDS(BeiDou navigation system), GPS(global position system)and their combination can continuously receive simultaneously, and positioning accuracy are obtained. The analysis shows that the GNSS orbit determination meets the development requirements of the new generation of geostationary meteorological satellites.

Aiming at the problems of large changes in the scale of aerial images of unmanned aerial vehicles (UAV), great recognition difficulties, and generally small targets, the paper proposes an object detection algorithm for UAV aerial photography based on improved SSD RCBnet. In order to improve the feature extraction capability of the network, the algorithm modifies the feature extraction network of the SSD algorithm to Resnet 50 and adopts the feature fusion method to fuse the feature maps, and uses the fused feature maps to build a feature pyramid. In addition, in order to enhance the algorithm's ability of objects detecting, a multi scale convolution structure of attention mechanism is designed to effectively adjust the receptive field and realize the parallel operation of the feature map of the convolution kernel of different sizes. Aiming at the problem of extremely imbalanced positive and negative samples in training, the algorithm uses the Focal Loss function to train the network model so as to focus on difficult samples. Compared with other classic algorithms, the algorithm proposed in the paper has higher detection accuracy, better detection performance and robustness in UAV aerial images. Compared with SSD, the accuracy is improved by 3.46%.

This paper studies the inverse consensus of multi agent systems with directional constraints. It is assumed that the agents of the multi agent systems follow unidirectional motion in one dimensional space. For the first order integrator model, a control input is designed, under which the inverse consensus is achieved. For the second order integrator model, a method to design the protocol control is proposed, under which the agents stop and reach a configuration in which every agent is located at a distance no less than the sensing radius from every other agent. Further, the inverse consensus of multi agent systems in multi dimensional space is studied. It is assumed that the agents of the multi agent systems have constraints in a certain direction. For the first order integrator model and the second order integrator model, a control input is designed, under which the inverse consensus is achieved. Numerical simulations have demonstrated the effectiveness of the approach proposed.

Aiming at the control problem of nonlinear underactuated unstable systems, a virtual full actuated control method based on auto coupling PID control theory is proposed. This method will define the known or unknown internal dynamics of each input channel as a total disturbance, and in the first input channel is introduced into the corresponding virtual control, thus the nonlinear underactuated equivalence of unstable system for virtual full drive system of linear disturbance, and based on the auto coupling PID control theory to establish the first virtual controller, input channel to form virtual instruction, then the controller of the next input channel is established according to the obtained virtual instruction and the auto coupling PID control theory to realize the effective control of nonlinear underactuated unstable system. The simulation results verify the effectiveness of the proposed control theory and have wide application prospects in the control field of nonlinear underactuated unstable systems.

As a mature modeling technology, the block diagram is widely applied. MATLAB/SimMechanics use this technology to model and simulate multibody mechanical systems. However, SimMechanics is obviously insufficient for flexible and large multibody mechanical systems requiring high modeling precision and real time simulation. While DARTSM software package can perform high precision and real time simulation of spacecraft mechanical system. This paper deeply analyzes underlying modeling technology of the two softwares, then designs the detailed method for model conversion according to the equivalence relationship of modeling information, and finally develops the model conversion program through C++ to convert the SimMechanics model of spacecraft multibody systems to DARTSM model. Through the model conversion method and program in this paper, the advantages of the two different softwares can be combined. So that SimMechanics indirectly obtain the ability to handle flexible body and real time simulation. Similarly, it also resolves the problem that DARTSM software package lacks graphical modeling interface, which has important engineering application value.

The libration points L4 and L5 of Sun Earth system provide an ideal view of the solar coronal mass ejections (CMEs) and corotating interaction regions (CIRs). A mission in the Sun Earth L4/L5 libration point orbits and solar polar orbit is proposed in this paper. The attitude control system for the Sun stereoscopic exploration is designed, and numerical simulations demonstrate that the pointing accuracy and stability can satisfy the high precision and high stability requirements of the sun pointing control.

The spatial parallel between the receiving and transmitting optical axis of the atmospheric lidar system directly affects the validity of the detection data. The high precision optical self alignment system is one of the guarantee conditions for the reliable detection of the space borne lidar. Based on the existing international space borne atmospheric detection lidar, the research status, basic principle and common methods of the optical path adjustment system are summarized. The common echo signal intensity method and spot adjustment method are compared from two aspects of hardware structure and software algorithm. The structure of echo signal intensity method is simple, but the self alignment process is significantly affected by atmosphere and surface environment. Spot self alignment method can achieve 1μrad high precision adjustment, but the optical path is complex and the cost is relatively higher. In general, the optical self alignment technology involves many disciplines and fields. Combining the advantages of each discipline, it is the future development direction to design a simple structure but high precision and low cost optical path adjustment system.

For satellite telemetry data, imbalance of distribution and lack of anomaly tags are main problems. Therefore, usual anomaly detection method is not effective in this respect. An anomaly detection method based on timing generation countermeasure network is proposed. Preprocessed to eliminate the noise and outliers, the satellite telemetry data is used to train the generator and discriminator, so that the time dependence of historical data can be learned. Finally, the adaptive threshold method is used to determine whether the residual of generator is abnormal. The experimental results show that the anomaly detection method is effective.

The existing purely data driven spacecraft health monitoring technology cannot cover the area without measurement and control, and the telemetry data is limited, so that it cannot meet the needs of space missions with complex functions, diversified tasks, and long term orbit. A model and data hybrid driven spacecraft health monitoring system architecture is proposed. The system is based on a credible digital model, combined with on orbit real time telemetry data, to achieve full time, continuous and reliable status monitoring and status prediction of the spacecraft. In this paper, the design of the core technologies  is described in detail, including model implementation and correction, consolidation of telemetry data and model, and fault diagnosis. The application and verification of the framework during a certain type of spacecraft mission is further described.

In view of the problem of chattering when sliding mode approaches are implemented for tracking control of hypersonic aircraft during reentry, a tracking control method based on multi slide mode controller fusion is proposed in this paper. The method first describes the hypersonic vehicle reentry process using a kinetic model, and establishes the angle of attack, side slip angle, and tilt angle tracking control equations. Then the quasi second order continuous sliding mode controller and super spiral sliding mode controller are combined to realize the tracking control in stages and respectively weaken the influence of shaking vibration, which helps to improve the tracking control performance during reentry. The paper verifies the effectiveness and feasibility of the proposed tracking control method through model simulation and shows that the control method can not only realize the track tracking of hypersonic aircraft, but also improve the performance of control torque response and attitude angle speed tracking error integration, which can effectively inhibit the chattering and improve the flight stability.