UAV swarms are a real threat on the current battlefield. Their low cost, high flexibility, and cluster combat characteristics have brought huge challenges to air defense interception. Using swarms to defend swarms is a critical means to counter UAV threats. This paper analyzed the combat process of swarm-to-swarm battle and concluded that the collaborative situational awareness capability of intercepting aircraft was the core step to achieving swarm-to-swarm intelligent interception. Collaborative situational awareness for swarm-to-swarm interception can be used to solve the problems of image registration and multi-target tracking under conditions of few available information dimensions, large spatial error dispersion, and complex dynamic scenes. The development direction of related technologies in swarm-to-swarm interception scenarios is determined from a collective analysis of image registration and multi-target tracking technologies.

Recently, the proliferation of drones has driven significant advancements in anti-drone radar systems. However, radar signals produced by the small radar cross-sections (RCS) and low velocities of drones is challenging to detect, especially when obscured by background clutter. This leads to high missed alarm rates and limited detection ranges when using traditional signal-to-noise ratio (SNR) detectors. In this paper, a method transforming the micro-Doppler signals from drones into jet engine modulation (JEM) Doppler signals was proposed. Then, a signal-to-clutter ratio (SCR) detector specifically designed to identify these Doppler signals was introduced. Test results from both a microwave anechoic chamber and outdoor environments demonstrate that micro-Doppler signals are captured in the radar echoes of both quad-rotor and VTOL fixed-wing drones across multiple radar bands. Besides, the SCR detector outperforms the SNR detector, exhibiting better detection stability, higher detection probabilities, and lower missed alarm rates.

UAV cluster operations have significant advantages, including high combat effectiveness, strong battlefield survivability, and extremely high cost-effectiveness. Thus, they are widely used in combat tasks such as infiltration reconnaissance, deception and interference, saturation attacks, and regional blockades, bringing great challenges to traditional air defense systems. This paper has summarized the characteristics of UAV combat, analyzed the relevant target characteristics, and combined UAV cluster countermeasures technology to form an Anti-UAV system construction idea, contributing to the development of UAV cluster countermeasures.

With the significant improvement of the anti-access/area denial capabilities of adversaries, the traditional offensive weapons of the United States face critical challenges. To this end, the United States must attempt great importance to the development of manned/unmanned aerial vehicle collaborative combat technology, hoping to improve complex task management and precise cooperation capabilities using its artificial intelligence technology to deal with various threats in a highly confrontation-complex battlefield environment. This paper reviewed the concept, top-level planning and related projects of manned/unmanned aerial vehicle cooperative combat in the United States was commended and its development trend was evaluated. In the future, manned/unmanned aerial vehicle cooperative combat will develop in stronger comprehensive system confrontation, higher architecture flexibility, lower cost of system combat capability and faster generation.

Based on the background of multi-UAV cooperative reconnaissance of multi-point targets, a clustering algorithm using reconnaissance coverage was designed in this study. The path planning and control problem of multi-point targets was solved by employing the ant colony algorithm and consistency algorithm. Firstly, a clustering algorithm was designed to reintegrate the target points to resolve the problem of repeated reconnaissance of target points. Then, the UAV flight path was projected according to the integrated target clustering results, and the reconnaissance sequence was determined from the clustered virtual target points. Finally, the consistency algorithm was applied to complete the formation-keeping and path-tracking algorithm of the UAV formation so that the UAV cluster could maintain a certain formation and complete the reconnaissance task according to the reconnaissance sequence. Simulation verification was carried out using the above algorithm, and the effectiveness of the clustering algorithm, path planning, UAV control and path control were verified.

The target information from the high-resolution remote sensing images using unmanned aerial vehicle(UAV) platforms is critical to military action planning. However, the large size and mixed background information of high-resolution remote sensing images can cause a lot of false alarms and misidentifications in the target information detection results. In addition, due to the huge difference between the image and the target size, detecting small targets in a large number of background pixels must employ a higher accuracy detection algorithm. To solve the above problems, this paper proposed a tiny target detection algorithm utilizing the block information of UAV remote sensing images. The model first alleviated the processing difficulties and slow speed caused by large image sizes using block information and a small neural network structure. Then, the global attention mechanism was applied to suppress possible false alarms. Finally, the detector and the classifier information during the training process was exchanged to improve the capabilities of both at the same time. The proposed detection algorithm was verified on a large-size remote sensing image dataset. The results show that the number of false positives in the detection results is significantly reduced and the accuracy is greatly improved, demonstrating the effectiveness of the proposed method.

With the upgrading of conventional air strike targets and the emergence of new aerospace threats, the field of air defense and anti-missile is undergoing a profound transformation, catalyzing fundamental innovations in operational simulation technology. Firstly, this paper studied the major challenges of current air defense and anti-missile operational simulation deduction in different dimensions, such as data, models, decisions, system integration, and credibility. Secondly, it explores the conventional methods and cutting-edge progress of simulation deduction. Finally, the paper prospected the future evolution trends of air defense and anti-missile operational simulation analysis technology. This paper gives a reference for the development methods, which are deeply intelligent, high-dimensional, adaptive, highly interactive, and cognitively integrated.

In modern warfare, the air attack target has fast attack speed and diversified attack modes, posing a greater challenge to the ground-to-air short-range defense combat mission. Considering the assignment requirements of rapid decision-making and saturated interception of incoming targets in ground-to-air short-range defense operations, this paper established a mathematical programming model. Utilizing a weapon-target assignment problem with the objective function of maximizing the cost-effectiveness ratio and the constraints of weapon resources and minimum expected kill probability, a two-stage allocation strategy was proposed, and a genetic-auction algorithm was designed. In the first stage, the number of future targets equalled the number of weapon fire channels by replication operation. In the second stage, the auction algorithm was applied to resolve the problem accurately. Compared with the existing simulation experiments of the intelligent optimization algorithms, the algorithm proposed in this study can significantly shorten the chromosome length in the first stage, compress the feasible solution space, and produce stable solution results, thus providing effective decision-making auxiliary information for the first-line commanders quickly.

Given the great dynamic change in the modern air defense combat environment, the air defense missile weapon system requires real-time online target allocation. In this study, a real-time target allocation method for air defense missiles based on a genetic algorithm (GA) optimized back propagation (BP) neural network was proposed. Firstly, the optimization model of the air defense missile target allocation problem was established by employing the number of air defense missile weapons and damage probability, and the maximum damage effectiveness was set as the optimization goal. Then, the air defense missile target allocation framework based on the GA-BP neural network was constructed. The optimal weights and thresholds of the BP neural network were acquired using a genetic algorithm to optimize the BP neural network. The accurate and efficient allocation of threat targets to the current air defense missile was achieved from neural network prediction. Finally, the optimized neural network was applied for simulation analysis, allowing real-time target allocation under the battlefield situation, and verifying the effectiveness and practicability of the proposed method.

Reliable detection and accurate measurement of low-angle targets are major challenges for ground-based air-defense radar. While the digitalization of ground-based information systems and phased array radar, exploiting environmental information as a priori and combining advanced models/algorithms brings the potential to improve. To that end, the paper established a structural model of the array receiving data by precisely formulating the propagation path of multipath echoes from low-angle targets using the digital elevation information provided by the Geographic Information System (GIS). With the aid of the spatial and intensity distribution of scatters, heuristic optimization was employed to optimize the side lobe of the beam pattern, which improved the signal-to-clutter/noise ratio and the detection probability of weak targets. In addition, to match with the structural model, the subspace-based estimator was modified, enabling accurate and super-resolution elevation estimation without pre-estimating the complex multipath factor. As verified by simulation results, the proposed model and method are available for ground-based radar to reliably detect and resolve multiple low-angle targets and give accurate elevation estimates.

Visible images, synthetic aperture radar（SAR） images, and infrared images have distinct advantages, including high resolution, robust resistance to environmental interference, and clear visualisation of thermal objects, respectively. Utilising the combination of these three modalities, the key components of a warship can be accurately detected under all-day and all-weather conditions. However, existing publicly available warship datasets only consist of single-modality images. Thus, there is a lack of datasets that allow three modalities to be aligned in both time and space. To resolve this issue, this paper proposed a method for constructing a multi-modal warship image dataset through a virtual engine and multi-modal data generation model. The dataset comprised 5 055 images, with 1 685 visible images, 1 685 infrared images, and 1 685 SAR images, each with a resolution of 640×640 pixels. To allow accurate detection of the key components of a warship, this paper introduced a method via adaptive region localisation. Specifically, the reconstructed three modal images were utilised as input data. A multi-scale feature extraction neck module was employed to acquire features by exploring the information complementary of multiple modalities at different scales. To better locate the region of the objects, a regional adaptive localisation module was designed. Finally, the key parts of the warship were detected accurately. Experimental results on the constructed multi-modal warship image dataset demonstrate that the proposed method can increase detection accuracy. When compared to the benchmark model, the proposed method improves the mean average precision by 5.52% when the IoU threshold is 0.5.

A distributed acoustic tracking algorithm for underwater targets, employing the Cramer-Rao Lower Bound (CRLB), was proposed in this study to improve air detection for underwater preset unmanned platforms. Firstly, this algorithm employed underwater distributed sonar to capture the radiated noise of aerial targets and applied signal processing methods to acquire the target's azimuth information. Then, integrating the pure azimuth observations from multiple underwater unmanned detection platforms enabled distributed pure azimuth tracking of targets. Finally, the underwater distributed unmanned detection system was optimized using the CRLB to enhance the robustness of air target tracking. Simulation results demonstrate that, in typical air-to-sea mission scenarios, the proposed algorithm successfully achieves a tracking error of within 5% and presents real-time node adjustment capabilities.

Tactical missiles extensively adopt combined guidance technology to enhance performance, using air handover strategies as their key technology. The traditional method employs Monte Carlo offline statistics to obtain the handover probability at a single time and online adjusts the seeker’s direction and search strategy, which requires a large amount of simulation and can only be used for preset operating conditions. Considering the above, this paper proposed an online evaluation method for missile handover probability based on an error propagation model. Main error sources were selected to establish the propagation model. The model was simplified to realize online computation and beneficial to multiple and strong universality. Through simulation verification results, it’s found that the online evaluation method for handover probability can achieve a relatively accurate distribution of handover probability and pointing error under different ballistic conditions. In addition, the accuracy of the evaluation satisfies the requirements of engineering applications.

To investigate the motion characteristics of the fairing separation of aircraft in a low-altitude high-dynamic-pressure environment, a numerical simulation was applied using the CFD-rigid body dynamics coupling method. This study has revealed the flow structure and dynamic behavior of the fairing during the process. Specifically, focused on the asymmetric inflow conditions, the disparate opening speeds of the two fairing halves, and the variations in the detachment angle of the fairing. Results show that under asymmetric inflow conditions, the separation of one half of the fairing is suppressed, being detrimental to a safe separation. Increasing the detachment angle is beneficial in terms of enlarging the windward area of the fairing while accelerating its separation from the aircraft.

To develop gas-generating agents which are clean, high production, and have a low residue rate, the formulation screening and preparation process of gas-generating agents using PSAN were investigated. The combustion temperature, gas production, residue rate, flame morphology and safety of the formulation were tested. The results show that the optimum ratio of oxidant, binder and plasticizer in the formulation is respectively 40%, 30% and 30%. The combustion temperatures of PSAN/PBT/A3, PSAN/PBT/BuNENA and PSAN/PET/A3 are respectively 1 441 K, 1 611 K and 1 309 K, and the gas production is 55.2 mol•kg-1, 49.3 mol•kg-1 and 57.7 mol•kg-1. The condensed phase contents are respectively 6.75%, 7.25 % and 6.57 %. The combustion flame of PSAN/PBT/A3 is the brightest. The flame brightness of PSAN/PBT/BuNENA decreases with a large carbon smoke emission, while the combustion of PSAN/PET/A3 produces a large amount of carbon smoke and contains only a weak flame. The impact sensitivities of PSAN/PBT/A3, PSAN/PBT/BuNENA and PSAN/PET/A3 gas generating agents are respectively 34.4, 21.6 and 35.8 cm, and the friction sensitivities are low.