National Fire Control Symposium (NFCS) is a professional conference in the United States engaged in the research of fire control technology.By introducing the situation of 2021NFCS,it is considered that the newly proposed seven-phase kill chain in NFCS,namely,prepare and configure,surveil, detect,track,identify,engage,access and defend,represents the new thought of American fire control research,and it is considered that the seven-phase kill chain is different from the past six-phase kill chain after preliminary analysis,which includes combat preparation and combat process,gives a more comprehensive description of combat,and emphasizes the preparation and configuration of kill chain.NFCS defines the relationship between kill chain and fire control by giving the requirements of seven-phase kill chain for the successful implementation of distributed integrated fire control.By describing the technical implication of the seven phases of kill chain and making appropriate adjustments to the meeting sessions, NFCS establishes the direct relationship between the combat and the technology studied by NFCS,and promotes the development of fire control technology at the level of fire control research.

In order to meet the needs of IFF(Identification Friend or Foe) of air targets in joint operations,an integrated method of IFF of air targets is proposed,which combines physical and tactical information.Firstly,the basic idea of this method is introduced by referring to the experience of the US military,and then the information fusion strategy is described.Based on this strategy,the homogeneous information fusion, heterogeneous information fusion and multi-period information fusion of the method are modeled by using evidence theory and intuitionistic fuzzy sets.Finally,the reliability of the method is verified by case calculation and comparative analysis.

To address the phenomena of incorrect segmentation and missing segmentation that occur when the current Deeplab v3+ model does not adequately employ high-resolution shallow features,an improved Deeplab v3+ feature image semantic segmentation algorithm that incorporates multi-scale features is proposed.In the backbone network,multi-scale pyramidal convolution is introduced.The standard convolution in the pooled pyramid of atrous space convolution is replaced by the deep separable convolution to reduce the number of parameters of the whole model.Finally,a multi-scale approach is adopted in the decoding layer to capture the global background,and the background features are combined with the shallow features and the atrous space pyramid pooling layer through the attention mechanism to enrich the semantic information of the fused shallow features.Experiments show that in CityScapes dataset,the proposed algorithm has a better edge segmentation effect,with an Mean Intersection over Union(MIoU) of 74.76%, which is 2.20% higher than that of the original algorithm.Compared with advanced algorithms,it is also proved that it is effective in improving incorrect segmentation and missing segmentation.

In order to solve the problems of slow convergence speed,easy divergence and uncertain model in under-driven AUV trajectory tracking,a dual closed-loop control strategy based on finite-time and reduced-order state observer is proposed.According to the principle of time scale,it is divided into position control loop and attitude control loop.The position control loop adopts finite time control method to speed up the convergence of position quantity.The attitude control loop adopts dynamic integral sliding mode based on the reduced-order extended state observer to quickly converge the attitude angle and compensate for the mixed uncertainties.The control effect of AUV trajectory tracking is simulated in 3D simulation environment.The simulation results show that the convergence speed,control accuracy,robustness and tracking effect of the proposed controller are higher than those of conventional trajectory trackers,and it can better meet the trajectory tracking control needs of under-driven AUV.

Aiming at the inconsistency of evidence in multi-sensor target recognition,a multi-sensor evidence fusion algorithm based on evidence belief entropy and similarity is proposed for target recognition.Firstly,a measurement model of sensor evidence uncertainty based on belief entropy is introduced by using the inconsistent uncertainty and nonspecific uncertainty of sensor evidence.On this basis,a method of generating sensor evidence weight based on confidence entropy and similarity is designed by combining the distance and conflict of sensor evidence.Finally,a multi-sensor evidence fusion model is constructed.The simulation results show that the proposed method is more effective than the traditional algorithm in target recognition.

With the characteristics of typical LPI radar signals and the requirements for modulation pattern recognition,a method of modulation recognition for typical LPI radar signal based on Gaussian smoothing ambiguity function and sDAE_LIBSVM is proposed.Firstly,the ambiguity function transformation combined with Gaussian smoothing is adopted to complete the construction of feature images.Secondly,a recognition network is built with the fusing of stack Denoising AutoEncoder (sDAE) and LIBSVM,which is used for the classification and recognition of feature images.The simulation results show that when SNR is -7 dB,the Probability of Successful Recognition(PSR) of the proposed method can achieve 97% for eight typical modulation patterns of LPI radars,including BPSK,Costas,Frank,LFM and T1~T4,and it has strong stability and robustness.Compared with other methods,it has better recognition performance.

Radar words are the primitives of radar phrase,and the extraction effect of radar words will directly affect the confidence of subsequent radar behaviour recognition.A solution is proposed for the radar word extraction problem in the case of unbalanced detection and acquisition data,i.e.an improved K-OPTICS radar word extraction algorithm based on K-means.By constructing virtual clustering centres and cluster merging,it achieves an extraction accuracy higher than 91.22% under various unbalanced sample simulation scenarios,and has better parameter insensitivity than the traditional algorithm does.

A UAV formation modeling and control method based on sliding mode controller with improved exponential reaching law is proposed.Firstly,the elastic system model is used to analyze the dynamics of UAV formation.Each UAV is regarded as a mass point,and the positional relationship among UAVs can be regarded as the binding force during formation control.On this basis,a fixed communication topology is adopted to analyze the balance state of UAV formation,and the dynamics model of UAV formation is established.At the same time,a sliding mode controller based on improved exponential reaching law is used to realize the cooperative control of UAV formation,so as to speed up the convergence to system stability,weaken the chattering phenomenon of traditional sliding mode controller and ensure the stability of cooperative control of UAV formation.Finally,the feasibility of UAV formation dynamics model and the stability of UAV formation controller are verified by actual UAV formation flight tests.

In order to realize the stable flight of six-rotor UAV formation under compound interference,a fast terminal sliding mode robust control method is proposed.Firstly,the topology of the leader-follower formation is described,and the 6 degrees-of-freedom motion model of six-rotor UAV is established.Then,the formation outer loop control law is designed to convert the formation command into the attitude command,and the rotor speed command is obtained by designing the formation inner loop control law.Finally,the adaptive law is introduced to estimate the compound interference,which realizes the stable flight of the six-rotor UAV formation.The simulation results show that compared with the sliding mode control method,the proposed method can realize the stable flight of six-rotor UAV formation more quickly,and the maximum trajectory error is only 0.2 m,and the maximum estimation error of compound interference is only 0.1 m/s2,which shows better control effect.

In a hazy environment,aerial photography equipment cannot accurately obtain image information.To solve this problem,an image dehazing algorithm with improved dark channel window and transmittance correction is proposed.Firstly,the hazy image is segmented by superpixel to obtain local windows with consistent depth of field,and the dark channel is calculated in each window.Meanwhile,the atmospheric light is estimated by using superpixel according to the characteristics of atmospheric light.Secondly,the transmittance is refined by guided filtering,and an adaptive tolerance mechanism is established to correct the transmittance of the bright area in the image.Finally,the atmospheric scattering model is inverted to restore a clear image.The experimental results show that the result of the algorithm is clear in detail and natural in color,and can handle multiple types of hazy images with better robustness.Compared with classical and novel algorithms,it has significant advantages.

In order to improve the aircraft trajectory prediction accuracy,considering route restrictions,cost index and aircraft performance restrictions,a multi-objective constraint-based aircraft descent flight trajectory prediction method is proposed.According to aircraft performance data,characteristic parameters,route restrictions and cost index requirements,dynamics theory is used to dynamically analyze and predict multi-dimensional state data for each flight sector.The take-off,climb and cruise data of a real trajectory in the aircraft data recorder are used to predict and optimize its descent trajectory.The simulation results show that,compared with the real trajectory,the predicted trajectory and multi-dimensional status data are dynamically adjusted with the cost index and the constraints of the route,which meets the route restriction and improves the fuel efficiency.

The closed-loop detection method for SLAM (Simultaneous Localization and Mapping) is prone to perceptual deviation in complex scenes with multiple ambiguities.Based on the closed-loop probability model,a closed-loop detection method that combines local SURF features with global ORB features is proposed.Firstly,robust SURF feature and global ORB feature are used to describe the image locally and globally.Secondly,the discrete Bayesian closed-loop probability model for multi-feature scene description is constructed,and the observation likelihood probability is constructed for multi-feature spaces,in which the local feature space calculates the observation likelihood probability based on the bag-of-words model,and the global feature space calculates the observation likelihood probability based on KNN nearest neighbor method.Finally,considering the temporal consistency of images,a multi-step,closed-loop candidate frame extraction method is designed based on epipolar constraints to further reduce the perception deviation.The experimental results show that the algorithm can eliminate most of the false-positive matching cases in multi-ambiguity scenes,and has better closed-loop detection effect and higher closed-loop accuracy compared with FAB-MAP2.0 and BoW methods.

With the continuous development of remote sensing technology,it is widely used in the fields of map drawing,resource exploration and disaster early-warning.Remote sensing target detection is the key step of remote sensing image interpretation.In the process of detecting remote sensing targets,the traditional detection algorithm has some deficiencies,such as target missing,low detection accuracy and inability to detect small target.A remote sensing target detection algorithm based on Multi-Scale Feature Enhancement Convolution Neural Networks (MSFE-CNNs) is proposed.By enhancing and fusing the features of different convolution layers,the model has faster training speed and higher detection accuracy.The proposed algorithm combines feature extraction module,feature enhancement module,self-attention mechanism and pyramid feature attention mechanism.The feature extraction module extracts features from the input of massive remote sensing data to obtain multi-scale features of different types of targets.The feature enhancement module is used for enhancing the correlation of features of different convolution layers and strengthening the learning ability of the model and the nonlinear relationship between features.Self-attention mechanism and pyramid feature attention mechanism mainly solve the problem that traditional convolutional neural network can not obtain the features of small-scale targets.To verify the effectiveness of the proposed algorithm,different algorithms are compared on DOTA data sets.Experimental results show that the proposed algorithm is superior to the existing target detection algorithms based on deep learning in both detection accuracy and training speed.

In order to solve the problem of high bit-error-rate of serial transceiver under strong channel attenuation,a low-power 112 Gibit/s SerDes transmitter is designed by using Duo-binary PAM4 coding technology.By adopting Duo-binary PAM4 coding technology,the problem of excessive attenuation of high-speed PAM4 (Pulse Amplitude Modulation-4) signal is solved.The system power consumption of the transmitter is reduced by using CMOS 1/4 speed architecture for 4∶1 MUX.The linearity of Duo-binary PAM4 transmitter is improved by using impedance calibration circuit.The transmitter is designed by CMOS 28 nm process and powered by 0.9 V voltage.The simulation results show that the transmitter can operate at 112 Gibit/s under the strong channel attenuation of 20.9 dB,with the power consumption of 1.9 pJ/bit and the linearity of 88.3%.

Aiming at the shortcomings of the traditional equipment test factor screening method,such as unreasonable test scheme design and low credibility of factor screening results,an equipment test factor screening method based on credibility is proposed.Firstly,the uniform experimental design method is used to construct a scientific experimental scheme,so as to obtain more experimental information with fewer experimental times and improve the experimental efficiency.Secondly,the absolute grey correlation method is used to calculate the correlation degree of the test factors,and arrange them from large to small;On the basis of the ranking of correlation degree,the research on the credibility of factor screening is carried out,and the credibility of factor screening is reflected by an intuitive measure.Finally,the rationality and feasibility of the method are verified by case simulation.

Aviation support equipment is the necessary basis for maintaining the combat capability of main battle equipment. In order to strengthen the construction of aviation support equipment system and improve the efficiency of aviation support equipment system,and aiming at the concept and characteristics of aviation support equipment system,an aviation support equipment system architecture modeling method based on activity-capability connection is proposed.According to the Department of Defense Architecture Framework (DoDAF)2.0,the aviation support equipment system architecture framework is constructed from the perspectives of support activities,capabilities,systems and services.Combined with the architecture design principles,the development process of view products based on this method is introduced.Finally,the feasibility of the method is verified by taking the information aviation maintenance support equipment as an example.

Aiming at the problem of synthetic aperture radar target recognition,a method combining multi-feature joint representation with adaptive weighting is proposed.The Principal Component Analysis (PCA), monogenic signal,and Zernike moment features are used to describe SAR images,and three corresponding feature vectors are obtained.Based on the joint sparse representation model,three corresponding features are jointly represented.The reconstruction error vectors from different features are fused using adaptive weighting algorithm under the framework of linear fusion.The optimal weights are achieved so the fused results can be improved.Finally,decision is made based on the fused reconstruction errors.Experiments are conducted on the MSTAR dataset for the 10-class problem under the standard operating condition,the conditions of noise corruption and partial occlusion,and the results verify the effectiveness of the method.

Aiming at the characteristics of initial trajectory of underwater high-speed target after coming out of water,a measurement method is proposed for synchronous observing with electro-optical compound device to guide the laser rangefinder and high-speed camera.A measurement device is designed,and the design requirements of each device that meet the general measurement requirements are analyzed,and the feasibility of the scheme is demonstrated.The trajectory measurement accuracy of the device is preliminarily estimated.The results show that it can meet the measurement requirements,which provides an important means for short-range measurement and live observation for initial trajectory of high-speed target out of water,and has great military application value.

The vector aberration theory shows that an optical system that has corrected the primary aberration will still have aberration performance in the state of small misalignment,which is particularly significant for airborne and spaceborne optical systems working in harsh environments.Wavefront coding technology has been proved to have a certain passivation effect on the first-order and third-order aberrations.Based on the principle of wavefront coding,this paper studies the passivation effect of the third-order phase plate on the astigmatism caused by the misalignment of the optical system from the perspective of vector aberration,and takes the Cassegrain system as an example for simulation verification.The research shows that the wavefront coding computational imaging system is not sensitive to the misalignment astigmatism caused by the tilt of components.It can maintain good imaging effect when the system is slightly misaligned.This has certain significance for reducing the difficulty of optical system installation and adjustment and enhancing the environmental adaptability of airborne and spaceborne optical systems.

By combining the artificial potential field method with ant colony algorithm,a path planning method of mobile robot based on artificial potential field and ant colony algorithm is presented.On the one hand,the influence factor of target point distance is introduced to improve the influence of potential field force on mobile robot path search.By improving the repulsion field function,the mobile robot is prevented from being unable to plan the optimal path due to large repulsion.On the other hand,constructing the potential field force heuristic function,taking into account the distance heuristic information and the potential field heuristic information at the same time,initializing the differential allocation of pheromones is conducive to improving the convergence speed of the algorithm.The experimental results show that compared with that of the algorithm in Reference ［15］,the optimal path length,the number of path turns and the convergence speed of the proposed algorithm has been improved by 2.6%,25% and 66.7% respectively,which shows the superiority of the algorithm in path planning.