Aiming at the difficulty of feature point matching in infrared and visible image registrationan infrared and visible image registration method based on NSCT contour extraction and main direction consistency matching is proposed.FirstlyNSCT transform is used to extract the contour curves of infrared and visible imagesand the corners on the contour curves are extracted as feature points by using CSS corner detection algorithm.Thenthe middle line of the contour curve is used as the main direction of the feature pointsand the SIFT descriptor of the main direction of the feature points on the infrared and visible images is calculated as the feature description of the feature points.Finallythe mismatched points are removed by using the method of main direction consistency matching of the feature pointsand the affine transformation parameters are calculated.Experimental results show thatthe average Root Mean Square Error (RMSE) of the proposed method is 2.36and the average Correct Matching Ratio (CMR) reaches 91.6%which ensures the accuracy of extracting and matching the same feature points on two kinds of images and improves the accuracy of image registration.

The basic Differential Evolution (DE) algorithm has the problems of insufficient search accuracy and convergence ahead of timewhich results in unsatisfying effects of 3D path planning.Thereforea modified differential evolution algorithm based on Sine Cosine Algorithm (SCA) is designed.Firstlythe mutation strategy is improved based on the search mechanism of SCA as well as the populations center of gravityand the crossover strategy is improved by integrating the disturbance strategyso as to improve the search ability and convergence performance of the algorithm.Thena new scaling factor is designed based on the Logistic functionso as to resolve the contradiction between global and local search.Through the experiment of function optimizationit is verified that the modified algorithm has good search accuracy and convergence rate.Finallythe modified algorithm is applied to 3D path planning of UAV.Owing to the advantages of the modified algorithm in searchingthe surrounding spatial environment can be better distinguished in each generation of searchso that the path selection is more reasonable.The simulation results show thatcompared with that of the basic differential evolution algorithmthe 3D path of UAV generated by the modified algorithm is shorter.

The single-stage target detection algorithm has attracted the attention of many researchers and industries due to its simple structure and high-efficiency model.Based on the existing YOLO algorithmregarding the difficulties of small size and tight arrangement of targets in remote sensing imagesa lightweight target detection method is proposed to improve the accuracy of small target detection in complex backgrounds.In this methodweighted fusion feature network is introduced to provide each layer of feature map with a weight coefficient that can be continuously learned in trainingso as to strengthen the feature fusion of deep and shallow layers.By introducing CIoU loss and model improvementthe convergence speed of the network is accelerated to meet the real-time requirements.Comparative experiments are carried out on the small target dataset based on DOTA remote sensing images.The experimental results show that the method has better detection accuracy and detection speed.

Aiming at the problem that traditional Direction of Arrival (DOA) estimation methods cannot process multiple sources with high accuracy while improving Degree of Freedom (DOF)a new two-dimensional DOA estimation method for MIMO radar based on the coprime array (SM-MIMO-CA) is proposed.This method is based on a new MIMO array-based coprime array model (MIMO-CA)which improves the accuracy of multi-source estimation when the number of elements is finiteand a larger array aperture is obtained with a smaller number of elementswhich improves the accuracy of two-dimensional DOA estimation.Finallythe effectiveness and reliability of the SM-MIMO-CA method in improving DOF and improving DOA accuracy are verified through experiments.

Aiming at the stability problem of stochastic nonlinear systemsan aperiodically intermittent controller based on discrete observations is designed to ensure the mean square exponential stability of the closed-loop stochastic system.Different from the existing intermittent controllersthis intermittent controller is aperiodic.Moreoverthis controller is a discrete-time state observation based feedback controllerrather than a continuous-time feedback controller.Thenthe criterion of exponential stability is givenand the intermittent controller is designed according to the criterion of stability.The mean square exponential stability of the stochastic nonlinear system is proved by using Lyapunov theoryand the correctness of the criterion is verified.Finallythe validity of the analysis results is proved by numerical examples.

Aiming at the instability and low convergence precision of evolutionary algorithms in UAV path planninga Refracted Sparrow Search Algorithm (RSSA) is proposed.FirstlyRSSA uses Refracted Opposite-Based Learning (ROBL) to expand the search range and develop a more concealed spaceso as to enhance the global search ability of the algorithm.Thenthe crazy operator is used to make the search of the algorithm more detailedand the convergence accuracy is improved.Finallythe algorithm of Simulated Annealing (SA)is integrated to refine each solutionso as to find a higher-quality solution.RSSA is compared with other five algorithms in six standard test functionsand the test results show that RSSA is reasonable and effectiveand the Wilcoxon rank test proves that RSSA has strong optimization ability.It is applied to the UAV path planning in complex terrainand the simulation results show that RSSA has the smallest variance and the lowest cost compared with other algorithms.

In order to make the midcourse recognition plan of ballistic missile more close to the actual combat sceneit is necessary to evaluate and analyze the plan.According to the actual combat scene and the existing detection capability levelan evaluation index system of the midcourse recognition plan for anti-missile early-warning is established,and the meaning of different indexes is explained.After thatexperts are invited to score the indexes of the plan to be evaluatedand the correlation degree between the indexes of the plan to be evaluated is calculated by using the method of matter-element theory.On this basisthe matter-element theory and the evidence theory are organically combined by using the normalized processing methodand the level to which the plan belongs is determined by using the formula of synthesis rules in the evidence theory.Experiments have proved that this method can effectively reduce the difficulty of obtaining mass functions in the evidence theory and their high subjectivitywhich makes the synthetic result more objective and accurateand the evaluation conclusion more convincing.

Regarding the problem that lightweight algorithm in remote sensing aircraft target detection is difficult to balance accuracy and real-time performancea model compression method based on YOLOv4 structured pruning is presented.In order to make the anchor frame parameters more suitable for remote sensing datasets and take advantage of network multi-scale detectionK-means++ algorithm is used to cluster the datasets and scale adaptive adjustment is designed to restrain the redundancy of the anchor frame caused by too many small targets and close target sizes.In additionin order to reduce the parameters of the modelthe scaling factor γ in the normalization layer is used for L1 sparse regularizationthe filter and convolution kernel weights are re-evaluatedchannels with less feature information are iteratively prunedand then the pruning model is fine-tuned to recover accuracy.The experimental results show that after pruningthe model parameters are compressed by 93.1%and the detection speed is 2.46 times faster than that of the original modelwhich can effectively improve the detection accuracy and real-time performance.

The detection and tracking of multiple small and weak targets is an important technical difficulty in radar detection.A new idea is proposed for the detection of small and weak targets in complex background.The detection performance is optimized from three aspects.Firstlytarget tracking in complex background is realized by improving the probabilistic data association algorithm.Secondlythe nonuniform clutter suppression under the condition of the squint array is realized by adaptive processing.Finallythe target tracking range is expanded and the target form is improved by analyzing the targets hidden information.Experiments show that this algorithm can be well applied to the tracking of targets with the feature of gate coincidencewhich can not only guarantee the accuracy of trackingbut also greatly improve the correlation speedand can be well used in the detection of small and weak targets in complex background.

Firstlythe factors that affecting the ammunition consumption of electro-optical countermeasure are analyzed.Secondlythe tasks undertaken by electro-optical countermeasure ammunition is calculated according to the factorsthe calculation model of electro-optical countermeasure ammunition consumption is established by combining the single task consumption of ammunition.Finallythe calculation method of the model is introduced with a concrete example.The calculation results show that the calculation model of ammunition consumption for electro-optical countermeasures presented is clearly described and easy to calculatewhich can provide a reference for assessing ammunition support in electro-optical countermeasures operationand has a strong practical significance.

Aiming at the nonlinearunderactuatedstrong coupling and multivariable control problems in trajectory tracking control of quadrotor UAVa cascade global fast Terminal Sliding Mode Control (TSMC) strategy based on Extended State Observer (ESO) is proposed.The control strategy enables the system to perceive promptly and eliminate internal and external disturbances in real timeand guarantees that the system will converge to the state of equilibrium in a limited time.In additionthe stability of the control system is proved by the Lyapunov theorem.The simulation results demonstrate that the proposed control algorithm can improve the robustness of the system and ensure the trajectory tracking accuracy of the quadrotor UAV under the conditions that the dynamic characteristics of the motor and external disturbances are taken into account,and the assumption of small disturbance of attitude angles is avoided.

National Fire Control Symposium (NFCS) is a series of professional conferences focusing on USAs fire control technology.The emerging topics in sessions and posts of NFCS in recent years are reviewed and classified.According to Call for Abstracts (CFA) of sessions in NFCSthe connotations of these topics are introduced one by one and then analyzed.NFCS is categorizingsummarizing and condensing these topics all the time according to the actual situation of the USAand has extended a few more noteworthy directions of fire control research in recent years.For examplerapid transition of new technology to the warfighterthe capabilities of fire control platformsand the application of autonomy in fire control systemsand so on.In additionthe USAs fire control research has displayed the characteristics of studying one topic from multiple perspectives.

Flocking system is one of the hotspots in the research of multi-agent systemand it is of great theoretical and practical significance to study its fission behavior.This paper summarizes the composition of flocking systeminformation interaction mechanism and stability analysis of fission.Firstlythe composition of flocking system is classified and introduced from the perspective of heterogeneous and isomorphic agentthe effects and limitations of fission methods based on isomorphic and heterogeneous flocking systems are discussed.Secondlythe information interaction of fixed neighbor distance and selective information interaction are summarized and analyzedand the shortcomings and application occasions of each information interaction mechanism are pointed out.Finallythe existing theoretical analysis methods of fission stability are introduced and analyzedthe future research direction of fission behavior is indicated.

The random error of MEMS gyroscope is a key factor affecting the accuracy of inertial navigationwhich restricts the development of inertial navigation.In order to improve the performance of MEMS gyroscope and improve the accuracy of Allan variance identificationthe random drift error of the gyroscope is analyzed by using the improved Allan variance method.On this basisthe random error model of MEMS gyroscope is established by using the time series analysis method.The results show that the method is simple in calculationflexible in modelingand can significantly improve the accuracy of Allan variance calculation and data utilization rate.The time series model is stable and has strong applicability.

In the process of image fusiontwo different-source video inputs often cause parallax between the two images due to the difference in optical system and spatial positionwhich in turn causes the problem of blurred edges and artifacts during image fusion.Thereforeimage registration is required before image fusion.Regarding the problem that the current heterogeneous image registration has poor real-time performance which makes it impossible to be applied in practical engineeringan image registration algorithm based on RANSAC and mutual information is proposedwhich divides the registration operations of different dimensions into the initial registration and real-time registrationin the meantime the POWELL optimization algorithm is introduced to speed up the searchwhich maintains a high real-time performance without reducing the registration accuracy.Finallythe proposed algorithm is transplanted to embedded hardware and accelerated based on Zynqand an embedded image registration system with high accuracy and real-time performance is realized.

In this paperdual-weapon multi-target bombing tactics is proposed in response to the operation mission requirements of bombing airports for gathering multiple targets.Firstlythe attack area of two small-diameter bombs with different ranges is calculated and simulated.The simulation results show that small-diameter bombs can maximize the combat effectiveness when applying at high altitude and high speed.Thenfrom the complete process of dual-weapon multi-target bombing tacticsmulti-target pre-grouping and re-groupingmulti-target attack conditions and multi-target attack direction optimizationthe theoretical method of dual-weapon multi-target bombing tactics decision-making is deeply studiedwhich lays a solid theoretical foundation for the engineering application of dual-weapon multi-target bombing tactics decision-making.

Ballistics simulation and launch envelopes simulation are key means of designing and verifying Air-to-Air Missile (AAM) performance.The two independent means of simulation are designed in architecture matched in interface and integrated in system.An integrated simulation system for AAM ballistics and launch envelopes is designed and validated.Experimental results show that the integrated simulation system can comprehensively complete ballistics and launch envelopes simulationwhich significantly improves the efficiency and quality for the designsimulationanalysis and assessment of AAM performance.

The study on friend or foe identification of air targets is seldom combined with the management of air battlefield.To solve the problema method for friend or foe identification of air targets based on fuzzy inference and evidence theory is proposed.Firstlythe correlation between the identification of air targets and airspace control of the battlefield is analyzedand the identification process and discrimination logic under the framework are introduced.Secondlythrough the design of an intuitionistic fuzzy inference systemthe corresponding inference rules and ambiguity-resolving algorithm are established to model the method.Thenthe evidence theory is used to modify and fuse the fuzzy inference results to achieve continuous recognition. Finallyan example is given to verify the validity and rationality of the proposed methodwhich can provide a reference for related research.

In communication reconnaissancemulti-channel Blind Source Separation (BSS) algorithm is commonly used to separate signalsbut sometimes the separation effect is not ideal.Through theoretical analysis and simulation verificationthe separation performance of common multi-channel BSS algorithms such as FastICAInformax and EASI in communication signal reconnaissance is studied.The influence of Signal to Noise Ratio (SNR)Signal Intensity Ratio (SIR) and relative Direction of Arrival (DOA) on multi-channel BSS is analyzedwhich provides a reference for BSS application research and algorithm selection.

For the issue that the distance-based approaches of connectivity preservation are too conservative a leader-follower formation control approach is proposed for second-order multi-agent systems with improved connectivity preservation.Firstlyconsidering that not all followers can receive information from the leadera desired offset position based distributed formation control protocol is designed.Nextby utilizing the position and velocity of the agents simultaneouslya judgment method of connectivity trend is proposed.Furthermorean adaptive action function based on relative velocity and distance is designed.Compared with traditional connectivity preservation methods based on artificial potential fieldthe proposed method can accurately determine the trend of connectivity between the agentsand reduce the adverse effects on the formation.Finallythe sufficient conditions for the stability of the system are given by using graph theoryand the superiority of the proposed method is verified through numerical simulation and comparative experiment.

Aiming at the operational application of outboard active bait on rainy daysthe influence of rainfall attenuation on radar detection performance is analyzedand the maximum radar operating range model under rainfall attenuation is established to obtain the relationship between the maximum radar operating range and rainfall rate.By analyzing the centroid jamming mechanism of outboard active bait and based on the factors influencing the jamming effect of outboard active bait in sunny weathera series of mathematical models under rain attenuation are establishedsuch as the equivalent Radar Cross-Section (RCS) model of outboard active bait and the missile tracking model.The shortest deployment distance of outboard active bait for effective jamming under different rainfall rate conditions is simulated and calculatedwhich provides a reference for the tactical application of outboard active bait.

In this paperthe control method of light panel auto-dimming in cockpit is studiedand a sliding mode control method for auto-dimming system is proposedthe design of control systemmodeling of controlled object and construction of sliding mode controller are completed.By using Matlabthe control rate is simulatedcomparing the system step response as well as the tracking accuracy with different parametersand optimizing the controller parameters.The results show that the control system has a fast response rategood tracking accuracyand can adjust the power consumption of light panel according to the change of the illumination of ambient lightthus achieving the control objectives of stabilizing the character contrastreducing the power consumption of light paneland improving the crews work performance.