Considering that it is difficult to locate the moving objects accurately from the clutter background by using the weighted kernel based color histogram to compute the feature of object template and candidate regions, we proposed a robust Mean-Shift object tracking algorithm based on improved weighted background to enhance histogram saliency.The saliencies of the object and of the background were calculated out from their bin of histogram, which was incorporated into the traditional similarity measurement for defining an improved weighted background coefficient.Therefore, the discriminabiltity of the object from the background was increased, and the effect of background information on object locating was reduced.Experiment result shows that the tracking effect is improved apparently.

The traditional method is not adaptable for threat assessment of beyond-visual-range multi-target combat of the fourth generation fighter, thus we proposed a new threat assessment method.Based on the traditional multi-target threat assessment method, the new method took combat height and stealth performance of the fourth generation fighter into account, calculated out the objective weight and subjective weight of threat assessment index respectively by fuzzy optimal method and IAHP, and finally gained the comprehensive weight.The improved TOPSIS was used for the multi-target threat assessment.An application example is given and the results show that the method is reasonable, effective, and suitable for the threat assessment of beyond-visual-range multi-target combat of the fourth generation fighter.

The rapid development of urbanization presents new requirements to autonomous collaboration capability of UAVs both in military and civilian application.Considering that the current mission planning algorithms for UAVs have the drawbacks of high complexity and poor real-time performance, we proposed a novel UAV mission planning method based on urban environment.Simple mathematical models were created for the urban buildings at first.Then, the improved visibility graph was applied to create all the shortest paths based on the models, and knapsack problem algorithm was used to allocate the missions.Simulations show that the algorithm can make an accurate planning that satisfies the real-time requirements.

Information warfare is characterized by System of System (SoS) operation based on information.From the viewpoint of operation information and combined with the OODA command and control loop concept, the network centric model of SoS operation was established by utilizing sensor node, information station node, decision node, communication node, lethality node and information jammer node.Based on the network centric model, the characteristic parameters of complex network were extended, and the information distance and interaction tempo of operation network were proposed.The effect of different topology and operation capability distribution on the interaction tempo was analyzed through experiment, which can be used in the research of SoS operation efficiency.

To obtain the best control performance for gliding Extended Range Munitions (ERMs), the real coding based adaptive genetic algorithm was combined with the best-keeping countermeasure, and corresponding improvement was made to the genetic operations.The improved algorithm was used for optimization of the control parameters.An improved ITAE valuation function was used to get the best dynamic response of the controller during the transition.Taking an ERM pitch angle controller as an example, the improved Powell algorithm, traditional genetic algorithm and the improved genetic algorithm were used respectively to optimize the control parameters.The numerical simulations show that the improved genetic algorithm has the best searching capability.The results are meaningful to the controller design for ERMs.

Aiming at precise ground-target attacking taken by UCAV, we proposed an optimal attack trajectory in the case with multiple constraints and a controller design method.A 3-DOF kinematics model was established for the missile, and transformation was made for the nonlinear equations thus no linearization is needed.An optimal attack trajectory with time and final angle constraints was designed by using the minimum principle of optimal control theory, which could satisfy the accurate real-time requirement.In order to track the optimal attack trajectory, a tracking controller was designed using sliding mode control based on the body mathematical model of a certain type of air-to-ground guided weapon.Numerical simulation results show that:1)The designed optimal attack trajectory can be achieved to implement accurate real-time attack under given constraints;and 2)The designed tracking controller has good robustness, high tracking precision, fine dynamic feature and stability.

Resource scheduling is an important part for joint operation, and reasonable combination of all kinds of operation resources will supply a guarantee to the operation.But during the operation, it is difficult to keep all the operation resources in good condition all the time.Once the resources can not be used, it will influence the accomplishment of the whole mission.In this paper, the Multi-Dimensional List Scheduling (MDLS) algorithm is improved and simulation is made, thus to reach the goal of resource rescheduling and guarantee the accomplishment of mission when certain resources are unavailable.

This paper is mainly focused on the research of magnetic attitude control algorithms for low-cost micro-satellites with only magnetic torque as the executor and no flywheel.First of all, a comparison is made to different attitude control algorithms for micro-satellites, then a new magnetic-only attitude control algorithm adaptive for engineering application is proposed.The new algorithm can improve the efficiency of satellite attitude control in its limited life cycle.At last, simulation is made based on the measured data and by introducing such disturbances as gravity gradient torque.The result indicates that the new algorithm can satisfy three-axis attitude stabilization of the satellite in short time, and has the advantages of high precision, light weight, small volume, low energy consumption, and easy for engineering application, which make it applicable for low-cost micro-satellites.

The kill probability of air-to-ground missile against complex targets needs to be researched systematically, and it is an urgently needed project to evaluate the kill probability in missile shooting test.A method for kill probability evaluation was proposed by taking air-to-ground missile against air defense radar vehicle as an example.The model for evaluating the kill probability was constructed based on the following five aspects:target vulnerability, power of warhead, model of the target, models of explosion point and fragment flying, and the comprehensive kill probability.The impact of CEP on kill probability was analyzed through Monte Carlo simulation.The result shows that the kill probabilities of fragmentation and the shock wave, and the comprehensive kill probability, are all closely relative to the changing of CEP.

In order to improve the robustness of missile guidance system to target maneuvering in the terminal guidance phase, a fuzzy variable structure guidance law was designed based on the variable structure control principle.Firstly, a variable structure guidance law was deduced through the theory of variable structure control.Then, the ideal guidance law parameters were found through simulations to reduce the chattering of guidance and enhance its robustness to target maneuvering. Moreover, the parameters of the guidance law were adjusted according to the target maneuver estimation by means of fuzzy control.At last, the designed terminal guidance law was compared with the traditional variable structure and proportional navigation guidance law. The simulation results show that designed guidance law has stronger robustness to target maneuvering.

Considering that the Wald Sequential Probability Ratio Test (SPRT) has poor real-time performance for detecting the soft failure of the airborne GPS receiver, we proposed a revised sequential Bayesian detection method.The method can modify the test statistics continually during the detection, thus can not only reduce greatly the testing time delay of the airborne GPS receiver soft failure, but also has minimum risks for decision-making of test.The simulation results show that:compared with the Wald SPRT method, the improved sequential Bayesian method can detect the soft failure of the airborne GPS receiver more quickly, which indicates that the method possesses a fine real-time performance.Furthermore, the method can decrease the probability of false alarm, so the airborne GPS receiver will have high reliability and stability.

Maximum Likelihood (ML) classification method is based on the assumption that the data are normally distributed, which is not always true for the realistic remote sensing data, and may result in decrease of classification accuracy.The classification results are impacted directly by the prior probability.The selection of training samples is somewhat stochastic and subjective.The ML method uses the same prior probability for the whole image, which will also reduce the classification accuracy.Theoretically, every smooth density function can be approximated to within any accuracy by such a mixture of normal densities.Thus the first problem of ML can be solved by using a combination of several normal functions instead of one.In this way, a very general capability can be provided, while still maintaining the convenient properties of the normal assumption.For the second problem, ISODATA is used to make a clustering image of the original data, after that, one can select the training areas of the image by comparing with the reference image.At last, the result of experiment shows that the proposed methods can not only realize the classification of remote sensing image but also achieve very high accuracy visually and mathematically in overall accuracy and Kappa coefficient.

A complexity based joint rate-distortion optimization algorithm is presented, which taking the complexity factor of video coding into account based on the rate-distortion optimization algorithm in H.264/AVC video coding standard.Thus a better effect can be obtained on the tradeoff between rate-distortion performance and video coding complexity.Besides that, this framework introduces a complexity control factor to control the extent of complexity.Tests show that the proposed framework can reduce the complexity without much sacrifice in coding efficiency.

The aviation users have high requests to integrity of GNSS.Based on the idea of improving the integrity level of GNSS combined system, an algorithm is put forward using protection threshold as an indicator for monitoring the integrity of GNSS combined system for aviation users.The algorithm integrates tightly the different satellite systems with different integrity monitoring methods, which can increase the aviation users' integrity level significantly.The simulation results show that the combined system of GPS/Galileo/Compass can provide global low-altitude flight and polar navigation with security assurance, and the effectiveness and feasibility of the method is proved.

It is the precondition of coordinated air combat to register the active sensors onboard multiple moving airborne platforms.Firstly, a biased measurement model based on WGS-84 (world geodetic system-84) coordinate system is constructed, and then the Maximum Likelihood Registration (MLR) algorithm is extended to active sensor registration on multiple moving airborne platforms in WGS-84.By using the chain rule for composite function derivation, the Jacobian matrix of sensor measurements with respect to target states is derived, which is critical for MLR application.Theoretic analysis and simulation results show that the method can realize sensor registration with the registration error approaching the Cramer-Rao Lower Bound(CRLB).

Time-to-go is an important parameter in guidance law design of air vehicles, and may determine the control gain and energy cost.This paper presents a new time-to-go estimation algorithm.Based on optimal control theory, the expression for terminal miss distance is obtained, and the correlation between terminal miss and zero-effort-miss is derived.To the intercept scene called Point of Closet Approach (PCA), a new closed-form solution to time-to-go estimation is given supposing that the evasive velocity is constant.The new estimation algorithm makes use of the included angle between relative position and relative velocity vector.Simulation result shows that:1) When the included angle is big, the PCA based method has less estimation error than traditional first-order approximation; and 2) The PCA based algorithm can also be applicable when the evasive velocity is variable.

A design scheme of a high-performance, commonly-used processing module is presented.The module selects four high-performance, dual-core CPUs to meet the data-processing and signal-processing requirements through parallel working, adopts three high-speed serial buses to meet the transfer bandwidth requirements, uses switch fabric to construct high-speed interconnect structure of the system with expandability and fault-tolerant/reconfiguration functions.The design of the module is in accordance with VPX (ANSI/VITA 46) standard, which can meet the requirement of the multi-task, real-time processing in the synthesized embedded system.

The problem of H∞ state feedback controller design is investigated for polynomial fuzzy systems based on sum-of-squares (SOS) optimization.Firstly, an H∞ performance criteria of polynomial fuzzy systems was proposed by using Lyapunov approach and SOS decomposition.Then, the sufficient conditions for the existence of state feedback H∞ controller were derived in terms of SOS based on the proposed performance criteria.Furthermore, the optimal H∞ controller design problem was converted into a parameter optimization problem with SOS constraints.Both theoretical and numerical simulation demonstrated the effectiveness of the proposed design approach.

The traditional test to ARINC429 bus only inspects the response of output to the stimulation of input, and is unable to cover all kinds of abnormal test environments．In order to expand the test coverage and initiative, we adopted the method of fault injection for studying the typical fault modes and implementation methods of fault injection in cases of physical layer, electrical layer and protocol layer separately．The results of experiment verify the effectiveness of the fault injection method.

Off-axis reflective collimator has been used widely in capability testing of infrared imaging system.There are a lot of testing methods, and new methods are also appearing.Based on engineering experience, the testing environment composed of an off-axis reflective collimator is introduced at first.Then, characteristic frequency and angular diameter of the infrared target are analyzed.A test method is presented based on the definition of infrared performance indicators, and the test result is obtained through the method.The characteristics of the thermal infrared imager reflected by the test result are evaluated in the end.

The noise of cockpit voice information has high loudness, numerous types and wide frequency range, which has serious influence on performance of cockpit voice recognition.To solve the problem, an adaptive filter based on Least Mean Square algorithm was used for noise reduction, which could achieve the best noise reduction effect by adjusting the order and the step length of the filter.After that, the cockpit voice was pre-emphasized, framed, windowed and conducted Fourier transform, followed by extracting Mel-Frequency Cepstrum Coefficients (MFCC) and first-order differential cepstrum parameters as feature vectors.Finally, a Support Vector Machine (SVM) was designed for training and identification.The problem that the performance of cockpit voice recognition is poor under a low SNR was solved.The simulation results show that this method is obviously superior to the wavelet packet de-noising, and the recognition accuracy rate reaches 96.9231%.

A modular adaptive disturbance rejection backstepping controller is proposed for the nonlinear attitude tracking of satellite in the presence of uncertainties, external disturbances and saturation constraints, which combines the theory of Nonlinear Extended State Observer (NESO) with constrained backstepping.Firstly, the satellite attitude is represented by Modified Rodrigues Parameters (MRP).And the NESO with parameter projection is employed to estimate and compensate the time-varying total disturbances on line for improving the robustness of backstepping controller.Then the backstepping controller is derived by introducing command filters and modified tracking errors to implement any operating constraints and obtain the derivatives of virtual control easily, which can also enlarge the projection sets of the disturbance estimation laws.Lyapunov theory proves that the closed-loop input-to-state stability is guaranteed under the effect of nonlinear damping.Comparative simulations state that the proposed controller has better performance in achieving precise attitude tracking and estimation of disturbance than classical adaptive backstepping.

Permanent Magnet Synchronous Motor(PMSM)is finding wider application in electromechanical actuation system, and has higher demands to the fault diagnosis.Therefore, we proposed a new method of fault diagnosis based on Local Mean Decomposition(LMD)and improved frequency-band energy method.This method unfolds the energy distribution of the fault signal in frequency domain by decomposing it into amplitude and frequency modulated signals.Then the feature vector is constructed according to the energy distribution and the correlation between each frequency band and the fault signal, thus to realize feature extraction.Finally, simulation is made for the speed control system of PMSM.The results turn out that the proposed approach can extract the fault feature effectively, while achieving dimension reduction of feature vector.

This paper introduces the principle of adaptive beam-forming technique, and presents a mathematical deduction to the algorithm in detail.A Global Navigation Satellite System (GNSS) anti-jamming method based on the adaptive beam-forming technique is proposed.Taking the circular antenna array with seven units as an example, the response of array is deduced, and a method for solving covariance matrix of the received signal is given.Therefore, the adaptive beam-forming technique can be used in the anti-jamming system of GNSS.The results of the simulation show that:this method keeps a gain with certain level in the direction of satellite signal and forms deep null in the direction of jamming, thus can have a good anti-jamming effect.