A set of long-wave infrared continuous zooming optical system was designed for airborne photoelectric detection equipment,based on long wavelength 320×256 element cooled thermal IR Focal Plane Array(FPA) detector.The optical system is made of two kinds of common infrared materials,Ge and ZnSe.By introducing a diffractive surface and three aspheric surfaces,the chromatic aberration and the off-axis aberration are well corrected, and the Modulation Transfer Function (MTF) of the system is kept above 0.35 for all continuous zooming ratios at the spatial frequency of 16 lp/mm.The system is consisted of 6 lenses,has a F/number of 3,working wave band of 7.7~10.3 μm,and cold shield efficiency of 100%.Continuous zooming from 50 mm to 400 mm is realized with smooth zoom paths.The results show that the compact structured continuous zooming system has pretty good imaging quality.

According to development trend of air-to-air missiles to multi-tasking attack,we proposed an air-ground integrated fire control and attack technique.Analysis was made to the process of air-ground integrated combat.Then,an adaptive guidance law for missiles was established based on the traditional air-to-air and air-to-surface guidance modes,which is applicable to fire control of air-ground multi-tasking integrated combat.Taking the current performance of missiles,the relative situation between missiles and targets,and targets value into consideration,a task-switching decision-making function was established based on the TOPSIS,to make the missile attack the optimal target.The simulation results show that with this guidance approach,the missile can adaptively select the optimum guidance mode in the attacking process,and complete the guidance to the specified target based on the mission requirements.It can also make the missile change the target in its flying process.

The feature descriptor of Pyramid Histogram of Words (PHOW) was studied,and an aerial reconnaissance image classification method was proposed based on PHOW feature descriptor and Support Vector Machine (SVM) classifier.The experimental results on five kinds of aerial reconnaissance images show that,compared with the image classification method based on Gist feature descriptor,the new method is able to realize image classification rapidly and precisely.The average classification precision is about 84.6%,and the time used for feature extraction and classification is about 37 ms.

To the problem of how to design a multi-UAV formation keeping optimization model,we combined the kinematic equation of multi-UAV,performance index and constrained condition together to construct a constrained optimization problem.After transformation was made to the performance function to decouple the optimization problem,we introduced the Lagrange multiplier vector to construct a Lagrange function.To the two classes of optimization variables,the basic and dual variables,we found that all the basic and dual variables could be formulated to the expressions of one dual variable.Then the gradient projection method from the convex optimization was proposed to solve this dual variable.And the other optimization variables could also be obtained through simple substitution.Finally,the effectiveness of the proposed strategy was confirmed by the simulation example results.

The issue of accompanying antisubmarine executed by multiple antisubmarine helicopters in naval ship formation ferry process was studied.Based on analysis to torpedo firing distance,we established torpedo attack position models for ideal condition and formation evasion maneuver condition,and obtained the antisubmarine configuration position.Then,the multi-aircraft accompanying antisubmarine methods were analyzed,and models were built up to calculate detection probability and the needed antisubmarine force.Finally,simulation was made to analyze the submarine torpedo firing position,and the effect of antisubmarine helicopter quantity,detection times,antisubmarine area size on the detection probability.The results show that: 1) The appropriate quantity of antisubmarine helicopter is three;and 2) The accompanying antisubmarine area should be set at 20 n miles in front of the formation with the size of 20 n mile×30 n mile.

Excessive lags of aircraft could cause Pilot Induced Oscillations(PIO).The frequency domain method and time domain method were used combinedly for studying the relationship between excessive lags and PIO quantitatively.A simplified closed-loop aircraft-pilot system model was built up,and the aircraft-pilot system stability was analyzed.In frequency domain,the frequency characteristic of effective aircraft was investigated by using Bode plot and Nichols plot, and quantitative estimation was made to the relation of the effective time delay and PIO by using bandwidth criterion and phase rate criterion.In time domain,the influence of excessive lags was quantitatively studied by using pilot induced oscillations time history and pilot rating.The results can provide a quantitative reference for controlling excessive lags of aircraft.

Considering that the size of collision avoidance zone is usually constant and can not make full use of the space,we proposed a method for establishing the collision avoidance zone of UAVs dynamically in real time.The collision avoidance zone is established according to the constraints of UAV maneuverability,collision zone,the relative position and the relative velocity.The influencing factors of UAV and the intruder to collision avoidance zone are considered dynamically,thus the collision avoidance zone is established more accurately with smaller airspace,which can not only avoid the unnecessary maneuvering for collision avoidance,but also make more reasonable use to the busy airspace.The simulation results show the good performance of this dynamic modeling method.

According to the requirements to real-time performance and high accuracy of aerial target identification in modern air combat,and considering the evidence conflicts that may occur in evidence combination,we proposed an aerial target identification method based on BP neural network and improved evidence combination rule.The BP neural network was used to acquire Basic Probability Assignment (BPA) of each sensor to target category judgment,which was taken as evidence for making spatial domain fusion and time domain fusion to each group of evidences by using an improved evidence combination rule.Thus the result of target identification could be obtained.Simulation shows that the proposed method can solve the problem of evidence conflict and implement the task of aerial target identification precisely and reliably.

To the issue of maneuvering target tracking,a modified de-biased converted measurement Kalman filter was proposed based on Interacting Multiple Model (IMM-MDCMKF),which used a Modified De-biased Converted Measurement Kalman Filter (MDCMKF) in multi-model.The MDCMKF algorithm improved the converted error covariance matrix by introducing state estimated value,and effectively reduced the effect of measurement noise on the covariance.Then the modified converted error covariance matrix was utilized to implement the DCMKF algorithm for a target tracking scenario.Finally,Monte Carlo simulation was carried out,and the results show that the accuracy of algorithm is superior to that of the IMM-EKF algorithm and IMM-DCMKF algorithm.

Considering the shortage of the traditional image enhancement algorithms that it can only use space-domain or transform-domain processing strategy separately,a novel processing method combining space domain with transform domain was proposed.Firstly,the original gray images were decomposed into different frequency bands by using Gaussian/Laplacian pyramid transform.Secondly,gamma correction was employed to achieve contrast equalization in each band.Finally,the enhanced images were reconstructed from the equalized one by means of inverse pyramid transform.Experimental results indicated that:Compared with traditional image enhancement algorithms,the proposed one can not only increase the contrast of underexposure regions,but also effectively suppress the luminance of overexposure regions,and achieve a good visual balance.

The tracking by detection algorithm implements target tracking by detecting the target in each frames in real time.This algorithm aims to maintain an online training classifier,which intends to separate the target from the background for tracking by taking the samples from the background area as negative samples and from the target area as positive samples.But when the target was sheltered,or the shape of target changed in a large scale,how to sample and mark the samples accurately was critical for success tracking.A tracking by detection algorithm was proposed based on structured Support Vector Machine (SVM).Since the output of structured SVM can be very complex data structure,the position of the target was taken as the output of the structured SVM,which can overcomes tracking drift problem when the target was sheltered or the shape of target changed greatly.Experimental results show that the proposed algorithm has a good and stable tracking performance.

Based on analysis to the working principle and composition of airborne directional infrared countermeasure(DIRCM)system,the development of some new foreign airborne DIRCM systems in recent years is introduced,and the perspective of DIRCM system is presented.

In order to reduce the dependence of Inertial Measurement Unit (IMU) calibration on turntable,and decrease its requirement to turntable control precision,we proposed a rapid accelerometer calibration method based on Particle Swarm Optimization (PSO) algorithm by analyzing the traditional model-observation-principle based iterative calibration method. At first,an objective optimization function was established based on model observation principle,which was taken as the fitness function of PSO algorithm for realizing the connection between the calibration method and PSO algorithm.Then,the accelerator calibration scheduling program was designed based on the estimated parameters sensitivity function corresponding to the maximum observation information.Finally,simulation was made to the proposed method and Newton iterative calibration method.The results show that,the proposed method is superior to the traditional Newton iterative calibration method in feasibility and effectiveness.

The Time-Triggered Ethernet(TTE) synchronization strategy ensures high-quality synchronization.The research on modeling and simulation of TTE synchronization strategy is presented in this paper.We proposed a modular approach for modeling and simulation of TTE protocol,constructed a clock model with linear drift,modeled the protocol state machine and the clock synchronization as discrete events.Through the modeling and simulation,the performance of TTE synchronization under typical configuration parameters was studied.The simulation result verified that the clocks of TTE devices could be synchronized within microseconds.

A kind of Terminal sliding mode controlling method is proposed for hypersonic vehicle with modeling uncertainties and external disturbance.Based on hypersonic vehicle pitch channel control system,a new integral type Terminal sliding mode is put forward.By introducing the first order filter and combined with Backstepping method,the singularity problem is eliminated.The compensation of controller performance is realized by means of the disturbance observer.The stability of the system is proved with Lyapunov theory to ensure the improvement of the dynamic performance of the system.The performances and robustness are analyzed through hypersonic flight simulations.The simulation results show that: 1）The disturbance observer can follow the tracks of the disturbance rapidly；and 2）The Terminal sliding mode controlling method can guarantee the stability and high control precision under the conditions of nominal parameter and large-scale parameter perturbation.

Considering the problem of surplus torque disturbance existed in the UAV actuator electric loading system,we proposed a compound control strategy combining linear quadratic optimal control with PID control for realizing the load torque control of the actuator electric loading system.The state space model was established with skillfully selected state variables for the electric loading system,and the optimal control law was obtained,which suppressed the disturbance caused by actuator movement effectively.Simulation result shows that the control strategy has good control precision and anti-interference capability.Compared with the compound control strategy combining CMAC neural network with PID,which is commonly applied in electric loading field,the proposed method is simpler and has smaller amount of calculation,faster simulation speed and smoother output.

The low-cost attitude estimation algorithm for Sitcom-on-the-Move (SOTM) system is usually complicated,and easily affected by the maneuvering acceleration and the sideslip angle.To solve the problems,we proposed an algorithm based on the spherical simplex transformation Unscented Kalman Filter (UKF),which fused the information from micromechanical gyroscope,accelerator and GPS for estimating the attitude accurately.To improve the real-time performance,the filter used the spherical simplex transformation to reduce the number of sigma points for speeding up the calculation,reducing the computation load and providing better filtering performance.The GPS-measured velocity was used to compensate for the maneuvering acceleration,and the sideslip angle was used to further correct the maneuvering acceleration when the vehicle was turning.Experimental results show that the low-cost attitude estimation algorithm is feasible for attitude stabilization of SOTM in both pitch angle and roll angle with accuracy of ±0.5°.

When small tail-sitter aircraft makes a high maneuvering flight,its attitude will change greatly.The traditional transformation algorithms from quaternion to Euler angle cant calculate out all its attitude angles.Thus we made some improvements to the traditional algorithms,and proposed an all-attitude conversion algorithm adaptable for Euler angle from -180° to 180°.This algorithm avoids the concussion of traditional algorithms at transform boundary under disturbance,and can ensure the continuity of attitude changing.Simulation result indicates that the proposed algorithm is feasible.

An actuator failure diagnosis algorithm was designed based on multi-model parameter estimation to monitor the working state online and present the failure type as well as failure degree in real time.Firstly,the dynamic characteristics of the typical actuator failures were described and the corresponding models were built up for each one.Then some state observers and parameter estimators were designed for each kind of the representative failures to monitor the operating state of the actuators and present the failure information in time.A switching mechanism was used to choose the correct observer for the current failure.The simulation result shows that the failure models we established can describe the dynamic behavior of the fault actuator very well,and the switching scheme can switch to the right observer correctly in time to supply the most accurate failure information.

The specific application of FC-AE-1553 in aircraft/store interface is introduced.It can improve the communication rate of data interface from 1 Mb/s to 2 Gb/s to satisfy the need of mass data transmission in future weapon system.A Fiber Channel (FC) interface card was designed and an FC-AE-1553 verification system based on switched fabric topology was built up.The command control and file transfer functions specified in SAE AS5653 (high-speed 1760 interface) were realized with small network latency and fast response rate.The verification system will promote the application of FC-AE-1553 in aircraft/store interface and other aviation and aerospace fields.

In development of system software for UAV Ground Control Station (GCS),it is essential to realize parallel development effectively and deal with the constantly changing demand.The working principles and realization ways of plug-in framework are presented,and the design scheme of plug-in UAV mission planning software framework is given.The design idea of the application framework is described in detail,and some key technologies of the plug-in manager,user interface management and the internal message communication mechanism are discussed.The design scheme was applied to an actual project and realized all-plug-in mission planning software framework,which proved that the framework can significantly improve the efficiency of software development,and improve the quality of software product.