| Title | Keywords | ||
|---|---|---|---|
| Author | Authorship | ||
| Corresponding Author | Funds | ||
| DOI | Column | ||
| Summary | |||
| Timeframe | - | ||
| Title | Keywords | ||
|---|---|---|---|
| Author | Authorship | ||
| Corresponding Author | Funds | ||
| DOI | Column | ||
| Summary | |||
| Timeframe | - | ||
The accurate prediction of remaining useful life(RUL) of engines is fundamental and central to achieving effective engine health management, and it is crucial for ensuring operational safety and reliability. Current RUL prediction methods of engines face issues such as insufficient local feature extraction ability and poor sensitivity to features across different time scales, making it difficult to comprehensively capture both short-term degradation trends and long-term degradation patterns. To address these challenges, an engine RUL prediction method based on dual time-scale feature enhancement and attention mechanisms is proposed. The dual time-scale feature enhancement technique is employed to extract operational features at different time scales, thereby fully capturing the degradation patterns during short-term engine operation. Furthermore, an attention mechanism is introduced to weight the features at different time points, highlighting the critical moments and improving prediction accuracy. Experimental results demonstrate that the accuracy of engine RUL prediction under various operating conditions by the proposed method can be effectively enhanced. It is of significant importance to improve engine operational safety and reliability, reduce maintenance costs and lower the failure rate.
Aiming at the mismatch between traditional electromagnetic simulation and measured scattering characteristics in Inverse Synthetic Aperture Radar(ISAR) imaging of group targets, a group target ISAR imaging approach based on multiangle dynamic scattering center modeling is proposed. Traditional methods struggle to accurately describe the complex electromagnetic scattering characteristics of group targets due to factors such as occlusion effects between targets, multiple scattering, and dynamic attitude changes. It is employed a component segmentation approach to geometrically decompose cooperative targets, combining the Shooting and Bouncing Rays(SBR) method to calculate multi-view dynamic scattering fields, and construct an attributed scattering center model to extract the amplitude and position parameters of scattering centers. In the imaging processing stage, a translation compensation is constructed to correct the translation term. For the nonlinear range migration caused by the rotation of target components, the Keystone transform is introduced in the range-Doppler domain to achieve linear migration correction, thereby ensuring phase consistency of the scattering center trajectories. Theoretical analysis and simulation experiments demonstrate that the method can achieve imaging of conical structure group targets at different azimuth angles, providing technical support for solving electromagnetic mismatch issues.
Structural lightweight design is the eternal goal of aircraft designers. As a key component of aircraft, the lightweight technology of rudder/wing is related to the overall improvement of aircraft performance. The development and evolution of rudder/wing structure in different aircraft types are introduced, the characteristics and research status of materials commonly used in different rudder/wing structure areas are analyzed, and the application of structural lightweight technology and related optimization design theory used in the design of rudder/wing "light-load-heat protection" scheme are summarized. Then the research progress of lightweight design of high speed rudder/wing structure under multi-field coupling conditions are summarized, and the development direction and challenges of structural lightweight design are pointed out.
Fiber reinforced polymer/plastic(FRP) is recognized as one of the core materials for the lightweight and high-performance development of tactical missiles due to their high specific strength/modulus, designability of mechanical properties, and multifunctional characteristics. Typical applications of FRP in domestic and international tactical missiles are reviewed, including load-bearing structures, aerodynamic and thermal-resistant control surfaces, and wave-transparent and stealth components. The current research statuses of key technologies such as high-performance material system development, lightweight structural design, high-quality manufacturing processes, and multi physics field coupling simulation are summarized. Based on this, three major challenges faced by the application of FRP in tactical missiles, as well as the development trends of multifunctional integration and structure-function unification, intelligent manufacturing and precision forming, and multi physics field coupling simulation with performance optimization are presented. The research can provide theoretical support and technical path reference for the new generation of tactical missiles to achieve lightweight, long endurance, and high breakthrough performance.
A predefined-time convergence control law is investigated for the problem of missile attitude control. A sliding mode surface is designed based on the predefined-time convergence theorem, and a predefined-time convergence controller is also given. A first-order predefined-time convergence disturbance observer is designed to estimate the sliding mode surface and the system disturbance, which not only simplifies the expression of the controller, but also resolves the singularity problem arising from the derivation of the sliding mode surface. The predefined-time stability of the system is proven by using the Lyapunov theory, and numerical simulation verifies the enhancement of stability and robustness of the control system by the predefined-time disturbance observer. The proportional-differential tracking guidance law is added, and the simulation is conducted in combination with the designed attitude controller, which further verifies the control performance of the controller system after adding guidance law. Simulation results show that the proposed control law can achieve fast and accurate convergence of tracking guidance commands under the complex interference conditions.
In order to accelerate the formation of actual combat capability of heterogeneous intelligent ammunition cluster, the research on technology of heterogeneous intelligent ammunition cluster for urban warfare is made. The form and characteristics of heterogeneous intelligent ammunition cluster for urban warfare are described, and the problems of traditional ammunition in urban combat are pointed out. By analyzing the characteristics and applicability of intelligent ammunitions with different configurations, the typical combat scenario of intelligent ammunition cluster in the urban environment is constructed. By sorting out the complex needs in urban warfare, the future direction of intelligent ammunitions is indicated. The research results are helpful for developing the heterogeneous intelligent ammunition cluster technology, and promoting the intelligent ammunitions to become the main equipment for future urban warfare, which can be used for guiding the practical applications of intelligent ammunition cluster in urban.
In response to the rapid development of open system architecture technology and the capability agility upgrade of airborne weapon, the enlightenment on the evolution of US open system architecture for airborne weapon development is discussed. The development of US open system architecture is introduced from the aspects of technology development status, typical project application and problems to be solved.On this basis, the design principle of airborne weapon open system architecture is introduced from the aspects of hardware and software decoupling, modular design and standardization design, etc. The design concept of airborne weapon open system architecture is explored from the aspects of hardware sphysical layer, module connection layer, operating system layer and application software layer, etc.The key technologies such as system architecture modeling and validation involved in the airborne weapon open system architecture are summarized. The future development suggestion of airborne weapon open system architecture is proposed from the aspects of architecture standard specifications.
In view of the fact that there is a large amount of temporal, spatial and electromagnetic domain data in the field of unmanned cluster collaborative target situation analysis, and due to the high data dimension, multi-source heterogeneous information and limited data processing means, traditional analysis methods are difficult to effectively deal with the problems above mentioned. A method combining literature research and technical analysis is adopted to systematically study the situation awareness technology driven by knowledge and data hybrid. A systematic review of the cluster target situation cognition technology system based on the hybrid drive of knowledge and data is conducted, revealing the intrinsic mechanism by which it breaks through information fog and reconstructs cognitive logic through the combination of knowledge guidance and datadriven approaches. At the technical level, the advantages and limitations of key methods such as multidimensional representation modeling, cross-modal semantic alignment, cross-domain information fusion, and intelligent reasoning are analyzed in detail. Solutions are also proposed for problems such as low efficiency of heterogeneous information aggregation, weak cross-domain knowledge transfer ability, and insufficient credibility of fuzzy reasoning,to promote the evolution of situation awareness from "data-driven" to "knowledge and data hybrid intelligence-driven", providing technical support for intelligent situation analysis and decision-making in unmanned cluster all-domain operations.
To address the increasingly complex dynamic threat environments in modern battlefields and the challenges of traditional path planning methods in meeting the requirements for cooperative stealth penetration by multiple unmanned aerial vehicles, an approach to multi-UAV stealth path optimization based on an improved 3D A~* algorithm is proposed. A 3D dynamic environment model incorporating radar and no-fly zones is established. Four evaluation functions are constructed, including UAV mission execution fault tolerance, flight stability, path smoothness and path stealth performance. The weighted sum of these evaluation functions is defined as the multi-UAV 3D path planning efficiency function. On this basis, an optimization model for multi-UAV stealth path planning in 3D dynamic environments is formulated, with the objective of maximizing the multi-UAV 3D stealth path planning efficiency function, while incorporating constraints such as no-fly zones and UAV maneuverability parameters. A two-step solution method based on the improved 3D A~* algorithm is employed to solve the optimization model. Simulation results demonstrate that the proposed method significantly improves planning efficiency while ensuring path stealth performance. Compared to the benchmark algorithm, the proposed method achieves a 20% enhancement in global path planning performance and an 85% reduction in computational time, validating its superiority.
Multi-agent reinforcement learning(MARL) is recognized as an intelligent enabling technology for cooperative control of unmanned aerial vehicle(UAV) swarms, with its essential mechanism established through distributed interactions and self-learning capabilities. A comprehensive review is conducted on recent research progress of MARL in UAV cooperative control. The fundamental theoretical framework of MARL is systematically introduced, followed by an in-depth analysis of its applications in addressing critical challenges of UAV cooperation, particularly in task allocation, path planning, formation maintenance, and reconfiguration. Existing research achievements are categorized and evaluated through representative application scenarios, including cooperative target tracking and aerial combat simulations, where limitations of current MARL algorithms are critically discussed along with corresponding improvement strategies. Key challenges are ultimately summarized, encompassing four dimensions: low convergence efficiency of distributed algorithms, sensitivity to communication latency, incomplete theoretical foundations, and significant sim-to-real gaps in practical deployments. This survey is designed to serve as a reference for advancing both theoretical frameworks and engineering implementations in multi-agent cooperative systems.