Conventional air-to-air missiles with aerodynamic control surfaces have limitations in terminal interception against high-speed, highly maneuverable targets. At the same time, the presence of exposed fins also constrains the loading efficiency on fighter jets. As a feasible and efficient control approach, nose deflection control technology has gradually demonstrated considerable engineering potential. It is expected to provide valuable reference information for the development of next-generation missile control technologies. This paper first systematically reviewed the historical development and current research status of nose deflection control. Then, it provided a comprehensive analysis of research endeavors related to aerodynamic characteristics, configuration optimization, dynamic modeling, and control system design, as well as the mechanism design and simulation of nose-deflection-based missiles. Finally, this paper summarized the significant research progress and anticipated the future trends in the development of nose deflection control technology.

In recent years, ongoing innovation within the military sector across tactics, technological capabilities of military systems, and equipment technologies has resulted in future air combat environments where airborne platforms will confront increasingly significant threats. These threats will be characterized by attributes such as high stealth, high intensity, comprehensive coverage, and multiple layers. As a result, the requirement for active defense capabilities has become increasingly urgent. Various international initiatives have put forward numerous concepts and initiated preliminary technical research. However, practical application in real combat conditions remains a challenge. This article focuses on the active defense mission of airborne platforms in future air combat. Starting with the threats these platforms face, it investigated the key requirements for active defense capabilities, systems, and technologies. Combining these with the latest developments in military equipment technology, this study summarizes the typical characteristics of active defense missions for airborne platforms.

For the problem of SAR target recognition in occluded scenarios, this paper proposes a novel method called the multi-view mutual learning network. The proposed method is a mutual learning framework consisting of a multi-view complementary feature learning network and a recognition network, which aims to improve the recognition model's feature extraction ability through knowledge interaction at the feature level. Specifically, to enhance the recognition network's feature extraction ability, this study developed an attribute-scattering-center-guided hierarchical feature extraction method. In view of the implementation challenges with target features in occluded scenarios, a multi-view complementary learning method was employed to comprehensively characterize target features by leveraging complementary features across different SAR images at adjacent azimuth angles. Contrastive experimental results on the MSTAR dataset show that the proposed method performs well across varying levels of occlusion.

Aiming at the cooperative search problem of multiple unmanned aerial vehicles (UAVs) for dynamic and static targets, a cooperative target search method of multiple UAVs based on an improved differential evolution algorithm (Adaptive Multi-strategy Differential Evolution, AMDE) and distributed model predictive control (DMPC) was constructed. First, the battlefield environment of the mission area was described by uncertainty and the target prior probability. Based on environmental exploration benefits, target detection benefits, and obstacle-avoidance constraints, a fitness function model for collaborative target search was established. Then, the AMDE algorithm, integrating multi-strategy adaptive selection and a double-layer adaptive parameter adjustment mechanism, was designed. Combined with distributed model predictive control, the optimal search track sequence was determined to complete the cooperative target search. The simulation results demonstrate that this method can effectively increase the likelihood of target discovery and improve the timeliness of task completion, while also exhibiting good efficiency and robustness.

Addressing the issue that traditional command and control (C2) architecture is intricate in satisfying the requirements of modern air defense combat systems, this paper analyzes the functional prerequisites of the air defense weapon C2 system and proposes a three-tier collaborative distributed architecture comprising the "core layer, edge layer, and access layer," developed in accordance with the principles of edge computing. The hardware configuration and selection scheme for the air defense weapon C2 system were designed, and the development requirements for its operating system were proposed, providing innovative ideas and methods for the intelligent upgrading of air defense weapon systems.

An Adaptive algorithm of FEM-SPH was used in this study to numerically simulate the high-velocity fragment penetrating the panel with different materials, and experimental results calibrated the numerical results. The calculation error was within 15%, and the calibrated model could be used to predict phenomena or reproduce experimental results. The penetration characteristics of tungsten fragments into the Q235 steel panel were examined. The influence of the fragment's mass, aspect ratio, impact velocity, and the panel's thickness on penetration ability was analyzed. The research indicated that appropriate combinations of fragment mass and size could be selected within the threshold range of fragment kinetic energy during the design of warheads, thereby increasing the number of damage elements and the probability of perforation. The research could provide a basis for the design of relevant warhead models.

In intricate and changing atmospheric conditions, turbulence may cause infrared images to appear blurry, distorted, or jittery. These effects significantly impact the performance evaluation of infrared imaging systems. A critical consideration in infrared scene simulation is an area that has garnered increasing attention from researchers in recent years. To address this challenge, this paper proposes a turbulence simulation model for dynamic infrared scenes based on the refractive-index structure constant. First, using measurement data and an infrared radiation transfer model, a three-dimensional, real-time infrared simulation scene was constructed. Then, by modeling variations in the atmospheric refractive index caused by non-uniformities in air density and temperature, the blurring and jitter effects induced by turbulence were simulated, enabling the generation of high-quality, highly realistic turbulence-degraded images. This study introduces an innovative method for simulating turbulence effects in dynamic infrared scenes, offering both theoretical insights and data to support further investigation into how turbulence influences infrared imaging systems and how to potentially mitigate its effects.

This paper addresses the increasingly complex fault diagnosis requirements of UAV (Unmanned Aerial Vehicle) systems by proposing a data-driven multi-model fusion fault diagnosis method. A deep learning model library incorporating CNNs, LSTMs, and RNNs was constructed to perform modular fault diagnosis across four key subsystems: guidance control, electromechanical, power, and airframe structure. Experiments employed a self-built simulation dataset, and model performance was evaluated using metrics including the confusion matrix, accuracy, recall, and F1 score. The results demonstrate that all subsystem models exhibit robust diagnostic capabilities, with the power subsystem achieving the highest accuracy, followed by the avionics subsystem. This technical approach offers a systematic solution for intelligent UAV fault diagnosis and can be applied to fault prediction in other complex equipment.

This paper systematically investigates a measurement-based computational approach for the inversion and interpretation of target electromagnetic scattering characteristics, focusing on the issues of stringent site requirements, high costs, and the limited capacity to reveal the physical mechanisms of target scattering associated with conventional Radar Cross Section (RCS) measurement techniques. First, high-precision electromagnetic near-field data of the target were acquired within a limited distance. Subsequently, the target's far-field scattering characteristics were accurately reconstructed using near-field-to-far-field transformation techniques. Then, the transformed data were processed by a parametric scattering-center extraction algorithm to invert the macroscopic scattering response into a set of attributed scattering-center parameters with precise physical meanings. By comprehensively analyzing the attributed scattering centers as well as the Inverse Synthetic Aperture Radar (ISAR) image, the link between“data measurement” and “physical mechanism interpretation ” was established, providing practical support for potential applications such as target shape optimization, recognition algorithm development, electromagnetic characteristic modeling, and professional wargame simulation.

To bridge the semantic gap resulting from the transmission of external products via heterogeneous communication protocols, in conjunction with the operation of the joint test platform utilizing the built-in standard object model (SDO) subscription-publish mechanism, this paper introduces an intelligent association method and system for protocol template-object models founded on large language models (LLMs). The process took the XML protocol template and interface description model as input, and generated the intermediate representation through structured preprocessing. The local knowledge base was constructed under the retrieval, augmentation, and generation (RAG) framework to uniformly store SDO definitions, historical association pairs, and proprietary corpora. And through the “dual-channel index,” combined with sparse keyword matching and dense semantic vector retrieval, highly recalled candidates for protocol elements and SDO attributes were generated. Subsequently, a lightweight, high-performance large-scale reasoning model was adopted to perform semantic disambiguation and consistency verification on candidate pairs, and the optimal match was output alongside the prompt-word norms and rule constraints. For “strange inputs” such as abbreviations, pinyin, and mixed Chinese-English writing, a multi-agent diversion parsing strategy was introduced, significantly enhancing the robustness against non-standard expressions. The system ultimately automatically generated a standardized XML association relationship list, supporting traceable evidence fragment backlinks and threshold filtering to avoid low-correlation strong matching. The prototype software integrated the full-process modules of file parsing, knowledge retrieval, intelligent matching, and result export. In verifying typical protocol templates and object model scenarios, it demonstrates robust alignment capabilities and engineering availability across languages and naming systems, providing an efficient and scalable semantic mapping path for rapid access to external resources by the joint test platform.

In a semi-physical simulation test rig for infrared guidance control systems, a low-temperature nitrogen environment is required to emulate the cold background of space. To ensure that the distortion of the light beam emitted by the infrared target simulator during transmission in nitrogen remains within the permissible range of the experiment, this paper establishes a flexible cold chamber simulation model for a low-temperature nitrogen environment. After experimental verification, the internal temperature distribution of the freezing chamber under various conditions was determined, and its optical transmission performance was subsequently evaluated based on this. Results show that under the rated operating condition, with an inlet velocity of 0.05 m/s and an inlet temperature of 153 K, the temperature standard deviation within the beam is 1.48 K, and the peak-to-valley optical path difference (Rδ) is 1.23 μm. The Strehl ratio (S) was 0.965, indicating a minimal effect of low-temperature nitrogen on optical transmission. Inlet velocity has a more significant influence on optical transmission performance. When the velocity ranges from 0.02 m/s to 0.07 m/s, the Rδ and S vary between 0.47-3.43 μm and 0.845-0.978, respectively. Notably, when the inlet velocity decreases to 0.015 m/s, the S drops to 0.798, resulting in the system degrading into a non-ideal imaging state. Additionally, an increase in position angle caused by the rotation of the upper cryogenic chamber decreases the threshold for temperature standard deviation in ideal imaging systems. Compared to a 0° position angle, this threshold decreases from 3.53 K to 3.23 K.

Addressing the issue of low consistency and credibility of traditional interference digital models compared to indoor interference generators, this paper proposes an intelligent optimization technique for interference digital models based on indoor data verification. Through the development of a multi-domain loss function that incorporates temporal, spectral, and energy aspects as the verification metric, coupled with the implementation of a Bayesian optimization strategy for the intelligent tuning of the interference digital model parameters, the congruence between the interference digital model and the indoor interference generator has been attained at a level of up to 90%. This high-consistency interference digital model, combined with indoor data, can be used to construct digital interference scenarios.

Multi-agent training is widely applied in modern maritime joint air defence operations. When two opposing parties independently choose different learning algorithms, it is often necessary to ensure implementation consistency and to manage resource allocation. To achieve this, a cross-agent shared fully connected layer is typically employed as a common representational foundation. However, a static choice between “fully shared” and “fully private” structures fails to balance policy consistency with individual differentiation. At the same time, discrepancies in behaviour distribution and gradient statistics across algorithms may amplify negative transfer and training variance. To overcome the above challenges, this paper introduced adaptive layer sharing (ALS) across heterogeneous algorithms, enabling learnable gating mechanisms to dynamically weight between shared and private branches at every layer. In a small-scale, single-machine experimental setup, standardized and reproducible protocols were established to record and report game outcomes and compliance. When ALS was activated, the learned gating distributions and thresholded topologies would be extracted, creating an implementable and observable engineering baseline that provides a clear structural and metric foundation for future large-scale and multi-task evaluations.