This paper proposes a novel distributed multi-target fusion mechanism based on Generalized Covariance Intersection (GCI) and Average Marginal Density (AMD). There are several inevitable defects in traditional multi-sensor tracking methods. For instance, the track-correlation performance is sensitive to the setting of correlation parameters, and the calculated amount of track correlation increases exponentially with the number of targets, etc. To solve these problems, a robust distributed fusion method suitable for multiple targets is proposed. Firstly, we approximated the local multi-target posterior probability density as a product distribution with its AMD. Secondly, considering the unknown correlation between different radar nodes, the GCI fusion algorithm was employed to perform distributed fusion. Since the track correlation process was embedded in the track fusion process, the distributed fusion mechanism performed track correlation and track fusion at the same time. Finally, we derived the closed-form solution of GCI fusion with AMDs. The proposed fusion algorithm is implemented by using Gaussian mixture model, and the experimental result shows that its performance is superior to that of the traditional fusion algorithm.

If the missile is converted to attack another target once it has aimed at one target, the repositioning will have a very serious impact on maneuvering capability of the missile. Based on the basic principle of the firing direction change of the missile body, this paper establishes a model of missile firing direction change, presents a method to realize the change by rolling of missile body in the boost phase. The simulation results show that: 1) The model can achieve the direction change in boost phase of ballistic missile; 2) The designed simulation system has certain anti-interference capability, and rolling angle command tracking effect is fine;and 3) The missile can fly steadily.

The concept of system-of-systems contribution is studied. To overcome the shortcomings of current methods for system contribution evaluation, the evaluation index system of the system contribution of space-based weapons is developed through the study on offensive and defensive capabilities of space-based weapons and their influence on the operation results. The quantitative mathematical model is built for the index with clear mathematical meanings. An evidential reasoning method based on fuzzy synthesis is proposed to evaluate the index with uncertain mathematical meanings. A typical space-based weapon combat system is developed, and the combat process of space-based weapons is analyzed by simulation for different operation modes, thus to evaluate the contribution of the weapons. The simulation results show that the index system for the contribution of space-based weapons is effective. It can successfully evaluate the system contribution of space-based weapons for a given combat system, and the result is reliable.

The reasonable matching of the combat tasks with the platform resources is an important condition for maximizing the operational effectiveness of the combat systems. Considering the difference of cooperation ability between the platforms, the operational effectiveness can be improved greatly if the platforms with high cooperation ability are put together to execute a task. Firstly, the models of cooperation degree between tasks and cooperation ability between platforms are established. Then a task-allocation model is established taking the minimal task completion time, maximal self-cooperation degree within a task and maximal mutual-cooperation degree between the tasks as the objective functions. A hybrid approach is proposed based on Dynamic List Scheduling (DLS) and Chaos Discrete Particle Swarm Optimization (CDPSO) to solve the model, for which DLS is used to select the task needs to be disposed, then CDPSO is used to select the optimal platforms for the selected task. Finally, a simulation is provided to demonstrate the feasibility and validity of the model.

To solve the problem of the attack area of the new type guided bomb, the dynamic model of guided bombs is built under the constraint conditions. The performance function with a weighted combination of the lateral and longitudinal ranges of the guided bomb is introduced, and the attack area problem is turned into the optimal control problem. Then, Gauss Pseudo-spectral Method is used to solve it. The impact of the speed and Flight Path Angle (FPA) at the starting control point of the guided bomb on the attack area is analyzed. The simulation results show that: 1)Gauss Pseudo-spectral Method can effectively solve the problem of the attack area of the guided bomb, and the attack area of bombs is increased with the speed and FPA;and 2)When the trajectory angle is larger than zero, the increasing of FPA has a relatively small impact on the attack area.

In order to solve the control problem in the launch control process of folding-wing UAVs, which is caused by the disturbances of both the step torque of the booster rocket and the rotation torque of the wing rocket, the optimal-state-control algorithm based on the optimal quadratic form is used to calculate the time range of the control law accessing the flight control system to participate in the controlling process. The precise time of the control law accessing the flight control system to control the rudder is finally determined by simulation tests. Using the optimal control scheme, the flight control system can overcome the oscillation within the shortest time and recover the steady state. Simulations and experiments show that: the optimal algorithm makes the effects of the dual torque disturbances on the system be lower, successfully solves the control problem of folding-wing UAVs during the launch process, and ensures the high quality of the launch control and the flight control of UAVs.

A backstepping non-singular fast-terminal sliding-mode-control method based on Finite-Time Disturbance Observer (FTDO) was proposed for a class of high-order nonlinear systems with unknown complex disturbances. The unknown complex disturbances was quickly and precisely estimated by FTDO and compensated by the designed robust items of the controller. The backstepping method was employed to deal with the high-order nonlinear system, and the dynamic surface control method was combined with it to design the virtual controller. A non-singular fast-terminal sliding-mode-controller was designed to improve the convergence speed and control accuracy of the system. The tracking error of the system has been proved to be uniform and ultimately bounded by using Lyapunov theorem. Finally, the simulation results demonstrated the effectiveness and feasibility of the proposed control method.

UAVs must be able to detect obstacles during flight and avoid the obstacles precisely.In order to make the UAV safer in flight, a method of obstacle avoidance is proposed combining binocular stereoscopic vision with optical flow.The binocular stereoscopic vision is used to obtain reliable parallax value through edge index algorithm, to get the spatial depth information according to the parallax convergence angle, thus to identify the object distance.The SIFT based optical flow method can obtain the movement speed of the obstacle relative to the camera at each moment.In order to obtain more accurate position information more quickly, the stereoscopic vision and optical flow are combined together.The experimental results show that this method can effectively improve the efficiency and precision of obstacle avoidance.

In view of the underdetermined situation that the number of sensors is less than that of the source signals, we studied the Undertermined Blind Source Separation (UBSS) problem based on Compressed Sensing (CS). Starting with the mathematical models of UBSS and CS, the source signal that has sparseness property is transformed into the issue of sparse signal reconstruction in CS theory. Then, the signal reconstruction algorithm of CS theory in the Sparco framework was applied in UBSS to construct a two-step CS-UBSS algorithm model. The Restricted Isometry Property (RIP) characteristics of the model was proved in theory. The simulation results prove the feasibility and applicability of the algorithm model for voice signals and image signals, which providing a new way for the solving the UBSS problem, especially in such a case that the reconstruction algorithm in CS theory can be applied directly in the recovery of source signals.

The suppression of altitude channel error of single-axis rotary Inertial Navigation System (INS) is studied. At first, analysis is made to the stability of altitude channel in pure INS, and an error model is established by using specific force equation. Then, the altitude channel error equation under the condition of single-axis rotary INS is established, and the conditions for suppressing altitude channel error divergence are analyzed. It is concluded that the altitude channel error can be suppressed efficiently when the rotation axis is in horizontal state. Finally, simulation is made to the altitude channel error of INS, influenced by single-axis continuous rotation under the condition that there is bias in triaxial accelerometer. Theoretical analysis and the simulation results show that the altitude channel error can be restrained effectively when the rotation axis of inertial measurement unit is in horizontal state.

In order to realize the monitoring and analysis of communication instructions between User Application(UA)and Cockpit Display System(CDS) in ARINC661 cockpit display system, we studied the instruction filtering technology. An analysis was made to interactive instructions in the communication process, and the structure of the binary tree was used to obtain a filtering algorithm, which is efficient for data filtering. Ground tests showed that the filtering method is correct, and the designed instruction monitoring and display software is reliable and has good performance, which plays a very important role in the development of cockpit display system.

A shallow Convolution Neural Network (CNN) is designed to replace the fully-connected layer in the pre-training model. The CNN feature extracted by the pre-training network is taken as input image to the designed shallow CNN network. Compared with the fine-tuning method for the pre-training model, this method can better adapt to the aerial image localization tasks. In order to further improve the locating accuracy of aerial images, the Bi-LSTM network is added to the network at the CNN classification stage by use of the temporally-continuous characteristics of the UAV aerial image. Thus the features of multiple images can be taken as the criterion for the network classification. Experiments show that the accuracy of sequential image locating method reaches a stable level at around 0. 89, and is improved by about 5% compared with the single-image locating method.

To improve the real-time performance of target track pose estimation, an algorithm for establishing dynamic template library with reduced dimensions is proposed on the basis of Jerk-model-based track pose filtering and the guidance of filtering innovation. First, the three-dimensional target model is reduced to a two-dimensional model by use of the relationship between the azimuth angle and roll angle of the track. Then, the track pose is filtered by using Jerk model, and the maneuvering state of target is judged based on the innovation of pose angle filtering, which is used for guiding the establishment of the lower-dimensional dynamic template library in real time. The simulation results demonstrate that the algorithm is effective for dynamic template establishing, which can reduce the size of library,improve the efficiency of pose angle estimation, and keep a certain level estimation precision.

There are some limitations of existing methods for obtaining the key parameters of quad-rotor UAVs in the time domain. For instance, they are not adaptive to varied environment and need a large amount of computation, along with high costs and long operating time in wind tunnel test. This paper presents an identification algorithm combining UKF data pre-processing with CIFER algorithm. Firstly, the system model of the quad-rotor UAV under hovering conditions is built. Then, the measured data of the quad-rotor UAV platform is pre-processed by UKF to get rid of the noise, which is then applied to CIFER software package to identify the uncertain key parameters of the model. The result shows that the result obtained from identification of pre-processed data has a better tracking performance at the low-frequency stage, the curves fit well, and the key parameters have high credibility.

Aiming at the decoupling control of the attitude and position of the quad-rotor aircrafts, a hybrid control strategy combining Dynamic Inversion Controller (DIC) with PID controller was proposed. Firstly, the quad-rotor model was taken as a single rigid body, and its nonlinear dynamic model was built through Newton-Euler equation. Then, we used the DIC theory to design the inner-loop controller, and the PID control theory to design the outer-loop controller. And thus the decoupling control of the attitude and position was realized. Finally, step response and upward spiral curve tracking were used to verify the stability and effectiveness of the proposed control strategy in the simulation. Meanwhile, the control code was written into the flight control system. The feasibility of the control algorithm was tested through flight tests. The results show that the proposed control algorithm can strongly resist the uncertain model dynamics and external disturbances, and it can satisfy the requirements of the attitude control and position control for the quad-rotor aircrafts.

Diffraction grating is a key component in holographic waveguide helmet-mounted display system. However, in color display, incident polychromatic light comprised of red, green and blue wavelengths will diffract different angles. After several total reflections in the waveguide, the light of three wavelengths arrives at different locations at the output grating, thereby resulting in dispersion. Aiming at the chromatic dispersion problem in the holographic waveguide helmet-mounted display system and based on the characteristics of holographic waveguide system itself, this paper presents a multi-layer design of the waveguide for dispersion correction. According to the diffraction angle of the incident light with different wavelengths, waveguide of four layers with different refractive indexs was designed. The optical paths of different wavelengths were corrected to achieve the goal of dispersion correction. Simulations were carried out to verify the accuracy of the design. The results show that: the red, green, and blue light almost exit at the same position, presenting a highly overlapping color pattern. Therefore, the method can effectively solve the problem of the dispersion of the holographic waveguide helmet-mounted display system.

High speed, high reliability and low latency make FC protocol the first choice for the next generation of the avionics network. But the event-triggered message transmission form makes the network lack time certainty. To improve the certainty of message transmission, we designed a time-triggered FC terminal interface. At the same time, the key modules of FC IP core, PCIE interface, the memory chip controller and the management module of send buffer were also designed. Finally, by use of the development tool ISE and Modelsim, we tested the interface by applying incentives. The result shows that BER of the interface is less than 10-12, and the eye jitter is controlled at 87. 5 ps, which meet the requirements of design.

Optical axis stability is an important indicator of the detection capability of an infrared optical system. Due to the lack of effective means of evaluation at the design stage, the stability check of the optical axis relied too much on the fitting and adjusting process formerly. Taking an infrared optical system as the object of study, we proposed a method to locate sensitive mirrors and structural modes for the optical axis. Based on LOS equations, the Multi-Point Constraint (MPC) equation between the optical-axis offset and rigid-body displacement of mirror surfaces is constructed. And the offset law of the optical axis changing with the temperature and its response spectrum under the effects of random vibration are obtained. The method presented has great significance for the stability design of the optical axis for complex optical systems.

Influenced by the operating environment and working time, the radio frequency impedance load is constantly changing, which may affect the output power and the efficiency of the RF source. Therefore, a matching network is needed between the RF source and the load, which can track the changes of load impedance dynamically and make corresponding adjustments to matching network elements, thus to maximize the output power of the radio frequency source. According to the changing characteristics of the radio frequency source load, model matching network is chosen in this paper. The control law of the adjustable elements of the matching network is controlled by the change rule of Smith chart. The current and voltage signals are detected respectively by Roche coil and the capacitive divider, which are then input to AD8302 amplitude/phase detection circuit for obtaining signal amplitude ratio and phase difference. After analog-to-digital conversion, the signals are input to STM32. According to the characteristics of impedance and admittance in Smith chart, PID algorithm is used to control the stepper motor and the steering angle, thus a new closed-loop control system is realized. The experiment and simulation results show that the circuit can be quickly matched in 5 s, and the error is controlled within 3 degrees.

The identification of the pilot actions in the cockpit is of significant importance for the accurate calculation of the pilot's workload. At present, the research about action identification generally focused on special actions, such as walking and jumping, which is not applicable for the pilot inside cockpit. As the pilot actions are basically implemented by the hand, the pilot action identification can be considered as the pilot hand movement identification. A novel technique is proposed here for the identification of the pilot actions based on the depth information, which tracks the pilots hands at first, and then identifies the pilot action. In addition, an approach to activate the algorithm is also presented for automatic recognition of the actions. The results of the experiments show that the identification rate of the technique is about 94. 06%, indicating that it can precisely identify the pilot actions in the cockpit.

According to the communication requirements of integrated avionics system onboard the new generation fighter, a data processing module system architecture based on RapidIO network was studied. A communication protocol software adaptive for avionics application mode and based on ARINC653 partition operating system was designed and implemented, which has found practical application in engineering. Function and performance verification showed that it can meet the large-data-volume and high-performance communication requirements in avionic system applications. The communication protocol software is implemented by hierarchical architecture, which is feasible for cross-platform migration, software reusing, and the subsequent hardware upgrading.

Airborne Integrated Electromechanical Management System (IEMS) is a safety critical system, ensuring the aircraft normal flight and the normal operation of each function of the plane. Analysis was made to the IEMS of the current fighter aircrafts, such as EAP, F-22 and F-35. A model based IEMS architecture modeling and simulation method was proposed, a common model of IEMS was established, thus the modeling and simulation process of IEMS was obtained. A monitoring algorithm based on state machine and truth table, and a redundancy management algorithm were designed based Simulink. Finally, the IEMS of a certain type commercial aircraft was used for modeling and simulation, which verified the feasibility, accuracy and effectiveness of the modeling method.

Most of super-resolution reconstruction methods are based on the condition that the point spread function is known or is Gaussian fuzzy kernel. However, the real point spread function of low resolution image, caused by random camera shake, is not Gaussian function. To improve the quality of the reconstructed super-resolution image and make it close to the real scene, a blind image super-resolution method is proposed based on L0 norm sparse representation. First, the point spread function in original images is estimated by using the gradient minimization method based on L0 norm, which is then used for removing the fuzzy effect of image in the process of super resolution reconstruction. At last, the backward propagation algorithm is used to make the reconstruction super-resolution image close to the reality. The experimental results show that: compared with bicubic interpolation and multi-dictionary learning method, the proposed method can get more clear reconstruction effect, and the PSNR and MSSIM are improved. The method has been validated by reconstruction test of real pictures.