At present, the refined damage assessment of reinforced concrete structures subjected to penetration explosion mainly relies on numerical simulations and experimental data analysis, both of which suffer from the drawbacks of time-consuming processes and high economic costs. To address these issues, a novel digital-intelligent prediction model for penetration-explosion damage is proposed to simulate the entire dynamics of the penetration-explosion process. Based on the graph neural network and a multi-task learning paradigm, this model can achieve rapid and accurate predictions of building damage morphology and node failures under various penetration positions and explosion yields, with a single penetration-explosion prediction taking only 0.2 seconds. Specifically, during the penetration phase, the structural response field prediction error is less than 2.9%, and the node failure prediction accuracy exceeds 99.5%; during the explosion phase, the structural response field prediction error is lower than 6.5%, and the failure prediction accuracy surpasses 98.8%. The evaluation of 100 penetration-explosion damage scenarios takes only 30 s.The proposed digital-intelligent prediction model demonstrates excellent generalization performance and highly efficient, accurate prediction capability, thus providing a new tool for building structure damage assessment and safety protection design.

With regard to the new development trends of artillery technology with the support of informationized and intelligent technologies, this paper summarizes the technological progress of artillery in aspects such as dynamic kill chains, long-range high-precision strikes, anti-jamming reliable strikes, high-density firepower strikes, and data-driven intelligent firepower. Furthermore, in combination with typical battlefield applications and new directions of equipment development, it analyzes the positive effects of these advancements on enhancing artillery performance. The research indicates that modern artillery has achieved severalfold increases in performance metrics concerning closed-loop strike time, range, accuracy, and rate of fire. It has also laid the foundation for artillery to adapt to the requirements of intelligent battlefield confrontation in terms of anti-jamming capabilities in complex battlefields and self-sensing/self-controlling functions of artillery systems. On this basis, this paper further explores the future development trends of artillery technology and proposes cutting-edge technological directions such as new-quality launching methods, extended-range technologies, and group firepower capabilities for artillery systems. It also preliminarily analyzes the significant implications of technologies such as high-initial-velocity launching through hydro-oxygen combustion, cannon-launched scramjet propulsion, and intelligent firepower coordination for the new-quality development of artillery, providing a preliminary vision for the future evolution of artillery systems.

Accurate determination of extreme boundary conditions in the forward design of a fuse is a robust guarantee for its safety. The safety margins hold paramount significance in the fuze forward design, where environmental excitations, detonation control components, and other safety factors directly impact fuze reliability. To address these challenges, this study proposes a conceptual framework for the fuze safety margins and establishes an analytical evaluation method for the electromagnetic environment safety margins in the fuze systems. The method specifically addresses two key scenarios: (1) voltage, current, and energy safety margin calculations under known electromagnetic environments, and (2) transient/pulse electromagnetic conditions. Furthermore, the paper systematically elaborates on experimental principles for safety margin verification in fuze detonation control circuits, principles for test point selection, and methods for component parameter selection.These contributions provide references and support for the theoretical analysis and engineering application of fuse safety margins.

To address urban traffic congestion and the high demands for traffic collaboration, this study proposes a communication cooperation model-multi-agent deep deterministic policy gradient (CCM-MADDPG) algorithm based on a communication cooperation module (CCM). The core of this algorithm lies in the design of the CCM, which explores the deep-level relationships among neighboring agents through an embedding layer and an information extraction module. It dynamically weights and fuses neighbor information with an attention mechanism to avoid redundancy and information averaging. By embedding the CCM into the Actor network, agents integrate neighborhood information for decision-making, leveraging the locality principle to alleviate joint action space challenges while enhancing system stability and collaborative efficiency. Experiments demonstrate that the CCM-MADDPG significantly outperforms baseline models, such as Qmix and FRAP, on both synthetic and real-world dataset, exhibiting exceptional scalability and adaptability in complex road networks. Specifically, in the Manhattan road network, it reduces average delay by approximately 28.8% and 26.3% compared with Qmix and FRAP, respectively. The ablation studies confirm the necessity of CCM's information extraction and attention mechanisms. This algorithm provides an efficient solution for multi-agent collaborative control, with dual potential for urban traffic optimization and applications. Its core contribution resides in the successful integration of the CCM design with the MADDPG framework.

In the context of modern military operations, the network collaboration capability of distributed unmanned combat clusters is of vital importance. Focusing on such cluster collaborative networks, an inductive graph reinforcement learning framework is established, and a heterogeneous network disintegration method integrating inductive graph representation learning and deep reinforcement learning techniques is proposed. The inductive graph representation learning empowers the method with the ability to rapidly generate disintegration effects on the strategies of emerging networks, and by modeling network disintegration as a Markov decision process and combining it with deep reinforcement learning for solving, the disintegration efficacy is significantly improved while reducing the complexity of the problem. Experiments show that the inductive approach is able to transfer among networks of different scales and achieves a better balance between disintegration effectiveness and time consumption; especially in scenarios where node-attack cost constraints are introduced, it exhibits excellent disintegration efficacy and strategy scalability. This research has significant military application value in reducing the complexity of battlefield cognition and improving the timeliness of dynamic decision-making.

Deep transportation tunnels and coal mine roadways in high geostress and complex geological environments are prone to engineering disasters such as large deformation of surrounding rock, coal bump and rockburst. The traditional cable with insufficient deformation capacity and limited impact resistance cannot adapt to the nonlinear large deformation damage characteristics of the deep buried surrounding rocks. A new type of energy-absorbing anchor cable with a shrinkable tube structure is proposed. This design leverages the plastic deformation energy absorption properties of steel to allow for greater deformation while providing high-strength, stable, and pressure-yielding bearing capacity. To explore its impact resistance characteristics, a series of drop hammer impact tests were conducted on the core pressure-yielding component of the anchor cable under varying impact heights, drop hammer weights, and numbers of impacts. The test results show that the shrinkable-tube energy-absorbing anchor cable can absorb energy up to 103 kJ in a single impact, and the working resistance is basically stabilized between 240~280 kN, which has good impact resistance performance. A field application test was carried out in the roadway of 8302 mining face in Xinjulong coal mine. The results indicate that the shrinkable-tube energy-absorbing anchor cable can achieve 120 mm of pressure-yielding deformation without fracturing. This research provides a new technical approach for controlling large deformation and impact disasters in deep rock masses.

The penetration depth of earth-penetrating projectiles into layered protective structures is a key issue in protective engineering design. In practice, the projectile usually penetrates the target with an incidence angle, and the protective structure is typically composed of multi-layered materials. The relevant theories and empirical formulas for penetration depth are relatively complex, which are not conducive to engineering applications. Based on the motion equation of the projectile, the motion deflection law of the projectile during oblique penetration was analyzed and a safety-biased calculation formula for oblique penetration was presented. The concept of equivalent thickness was introduced to convert the equivalent thickness of multi-layered media through wave impedance, then a calculation method for the penetration depth of multi-layered media was obtained. The results show that the oblique penetration depth can be estimated by multiplying the normal penetration depth by the cosine of the angle of incidence, yielding a conservative result; the penetration depth in different media is inversely proportional to the medium's impedance. The derived formulas for oblique and layered media penetration depths are consistent in form with current specifications, elucidating the physical significance of the empirical formulas in the specifications. Compared with the oblique penetration test in reinforced concrete, the maximum relative error of the theoretical prediction is 21.6% in the range of incidence angle from 0° to 50°. Compared with the numerical simulation of penetration to layered media , the theoretical prediction of the number of penetration layers is consistent with the simulation, and the relative error of penetration depth is 3%, which verifies the applicability of the theoretical calculation method.

The mechanical behavior of structural planes at various scales within a rock mass determines the stability of the surrounding rock in engineering projects. With the continuous increase in the burial depth of underground engineering projects, the threat posed by time-delay engineering disasters under disturbed environments is becoming more and more prominent. Based on a systematic review of research from field observations, laboratory experiments, and theoretical models, it is concluded that the remote induction of tectonically controlled time-delayed rockbursts and engineering seismic events is essentially the delayed instability of rock mass structural planes triggered by disturbances.A new concept termed the disturbance aftereffect of shear creep on structural planes is proposed. This concept refers to the phenomenon that, when a structural plane undergoing shear creep is subjected to a disturbance, the influence of the disturbance persists even after the disturbance has ceased. As a result, the subsequent creep process of the disturbed structural plane exhibits different states and nonlinear behaviors compared to its undisturbed condition. Based on the typical manifestations of the disturbance aftereffect, the critical transition behavior of shear creep in structural planes and its influencing factors are explored. In addition, this study highlights that the scientific characterization of disturbance intensity, the conditions and mechanisms of delayed instability critical transition, theoretical models of the critical transition process, and cross-scale research of rock-structural plane-rock mass systems are key issues that urgently need to be resolved. The solution of these problems will provide a solid foundation for understanding the mechanisms of remote triggering of earthquakes and the effective prevention of time-delay engineering disasters.

To investigate the distribution pattern of shock waves in relation to soil damage characteristics under the action of air shock waves generated by ground explosions of TNT charges, static ground explosion tests with 10 kg and 20 kg TNT charges were carried out, and the characteristics of air shock wave overpressure distribution after the explosion of charge were obtained. A simulation model for the pressure field of ground explosion shock waves was established, and its reliability was verified with experimental results.The research findings indicate that a frustum-shaped crater is formed on the soil surface after the charge explosion . The increase in the explosive charge from 10 kg to 20 kg results in an increase in the diameter of the crater from 1.5 m to 1.9 m, an increase in the air shockwave at 5 m from 0.186 MPa to 0.366 7 MPa, and an increase in the reflective pressure on the surface of the wall at 5 m from 0.604 MPa to 1.485 MPa. The shape of the explosive charge has a significant impact on the near-field pressure distribution. The ground reflection effect substantially enhances the pressure values in the near-ground pressure field, with the pressure in the near zone exhibiting a convex profile that gradually degenerates into an ellipsoidal shape at a distance of 5 m from the explosion center. The wall reflection pressure of the shock wave decreases with height, following a quadratic function attenuation pattern.

To address the issue of the quadratic growth in simulation time with increasing numbers of Unmanned Aerial Vehicles (UAVs) for multi-UAV combat mission simulations on Central Processing Units (CPUs), a lightweight, GPU-parallel accelerated simulation environment for multi-UAV combat is designed and developed. Unlike existing simulations frameworks that are often limited to single-UAV missions or only parallelize multi-UAV dynamics, the proposed framework, focusing on multi-UAV adversarial scenarios, further implements parallel computation for the relative poses and damage relationships of each UAV within the environment. In response to the limitation that traditional Basic Fighter Maneuvers (BFM) are only well-defined in level flight, an improved set of fully-attitude interpretable basic aerial combat maneuvers is designed. The simulation environment includes a variety of scalable multi-UAV confrontation tasks, which can provide heuristic baseline strategies for various tasks and support proximal policy optimization-based baseline strategies for one-on-one confrontation scenarios. Compared with the multi-threaded simulations running on an 8-core, the 16-thread CPU, the proposed platform improves sampling speed by two to three orders of magnitude. The reinforcement learning training results in both CPU and GPU parallel environments demonstrate that the GPU-accelerated simulation can reduce training time to 1/50, without significantly compromising sampling efficiency due to trajectory fragmentation. The proposed parallel-accelerated multi-UAV combat simulation environment significantly enhances sampling and training speeds, thereby facilitating accelerated research progress in this field.

Ground penetrating radar (GPR), as an efficient and rapid non-destructive detection technology, is often constrained by environmental noise interference and difficulties in image interpretation during shallow subsurface detection. Particularly in the detection of landmines and explosive remnants of war (ERW) , the false alarm rate remains high. To address these issues, this paper reviews advances in shallow-depth GPR technology focusing on the four critical dimensions: hardware architectures, soil environment, landmine/ERW detection, and signal processing techniques. The technical analyses and experimental investigations were conducted on prevailing shallow subsurface detection challenges, with a special emphasis on exploring the application of deep learning techniques in GPR signal clutter suppression and target recognition. This work provides valuable insights for improving GPR detection efficiency and effectively reducing the false alarm rate, and is expected to offer all-weather technical support for the rapid clearance of unexploded ordnance and military engineering surveys.

To investigate the influence mechanism of liquid filling ratio on the blast resistance of liquid-filled cylindrical shells, this study conducts a systematic examination of the dynamic response of partially filled cylindrical shell structures under near-field underwater explosions through integrated experimental and numerical simulation approaches. The results reveal that the structural response demonstrates three distinct phases: shock wave action stage, stable stage, and bubble pulsation action stage, with the primary deformation occurring during the bubble pulsation action stage. A significant positive correlation exists between liquid filling ratio and blast resistance, evidenced by a 52.7% reduction in central deflection when the filling ratio increases from 0% to 95%. While the internal peak pressure during the shockwave phase increases progressively with elevated filling ratios, the maximum internal pressure during bubble pulsation occurs at the 90% filling ratio. The plasticity of cylindrical shells under explosive loading decreases proportionally with increasing filling ratio, exhibiting reduction trends consistent with central deflection. These findings provide theoretical foundations and engineering insights for blast-resistant design and damage assessment of underwater protective structures.

To solve the problem of low reaction rate and incomplete reaction of metal fuels, a core-shell structured composite energetic material, AP@TiH2, consisting of ammonium perchlorate (AP) particles coated with titanium hydride (TiH2) powder, was prepared by the solvent-nonsolvent method. The microstructure distribution and ignition combustion performance of this composite material were studied by characterization technique and laser ignition experiments. The results show that Ti is uniformly attached to the surface of massive ammonium perchlorate particles, and the AP@TiH2 composite material has brighter flame, larger reaction area, shorter combustion time and faster reaction rate. The addition of AP@TiH2 to solid propellants can effectively improve the ignition characteristics of the propellants, and improve the combustion performance. Compared with the traditional mechanical mixing methods, the temperature rise rate of AP@TiH2/HTPB propellant prepared by the solvent-nonsolvent method is faster, the average temperature of stable combustion stage is higher, the ignition delay time is reduced by about 10%, and the linear combustion rate is increased by about 11%. The coating structure reduced the mass and heat transfer distance between the fuel and oxidizer, and effectively improved the reaction rate and reaction completeness of the fuel. This fine control method is expected to provide a new way for the effective utilization of solid propellant energy.

This paper addresses the challenges in collaborative Simultaneous Localization and Mapping (SLAM) for heterogeneous robot teams in global-navigation-satellite-system-denied environments, including pose alignment errors, communication constraints, and accumulated estimation drift. A multi-stage cooperative SLAM architecture is proposed to overcome these issues. The solution features a two-stage relocalization mechanism that employs Scan Context for coarse alignment and the Generalized Iterative Closest Point (GICP) algorithm for fine registration, enhancing initial pose estimation accuracy across heterogeneous platforms. A lightweight communication framework leveraging high-speed network-address-translation traversal technology is developed, which utilizes point cloud voxel filtering and keypoint extraction for data compression, ensuring real-time transmission in low-bandwidth conditions. Furthermore,the Random Sample Consensus (RANSAC) algorithm is incorporated to optimize multi-robot pose transformations by eliminating outliers, thereby improving global consistency. Multi-source point cloud fusion is subsequently achieved through coordinate transformation.The extensive experiments in underground parking garages and utility tunnels demonstrate the system's robustness and real-time performance, successfully generating complete point cloud maps with relocalization times under 1.58 s, RANSAC optimization under 3.2 ms, and map merging under 1.8 ms. The quantitative results confirm that the collaborative mapping accuracy is improved by over 45% compared with single-robot SLAM, and the integration of RANSAC further boosts relocalization accuracy by more than 15%, significantly enhancing the overall mapping precision. The proposed method offers a valuable reference for multi-robot collaborative perception and navigation.