With the increasing complexity of aviation equipment and the transformation of maintenance modes， structural digital twin technology has become a key enabler for structural health management and predictive maintenance. Addressing common challenges in digital twin modeling-such as modeling assumption deviations， input uncertainty， and model response mismatch-this study proposes a residual-driven model optimization mechanism based on multi-parameter flight measurements. A dynamic closed-loop framework of “measurement–calibration–residual feedback–model correction” is established， with a rigorous theoretical proof of the residual feedback mechanism’s convergence and a quantitative analysis of error upper bounds. Furthermore， a multi-dimensional， quantifiable evaluation index system for model self-evolution is developed. Engineering verification， using the tail of a certain aircraft as an example， demonstrates that the proposed method effectively reduces model prediction errors under complex operating conditions and improves the accuracy and robustness of fatigue life prediction. The research outcomes provide theoretical support and methodological foundations for the engineering application and intelligent development of structural health management in aircraft.

By the virtue of its design orientation of “multi-functionality” and efficient repeated flights in large airspace and wide speed range， the cross-domain vehicle is of epoch-making significance in the field of aerospace and is becoming one of the focuses of technological competition among various countries. Due to the high frontier and comprehensive nature of this field， there is no clear technical route to realize the large envelope cross-domain vehicle， but it is generally agreed that the morphing configuration is the key technology and necessary means. In this paper， we discuss the conceptual routes to achieve cross-domain vehicle from four aspects， namely， the basic configuration， the morphing mode， the optimization of the morphing structure， and the thermal protection， with the morphing configuration as the main theme. We analyze the technology pavements and reference values provided by the existing researches on the design of cross-domain vehicle， and summarize the respective technological advances and challenges. Firstly， several basic configurations with application potential are introduced， and the differences in aerodynamic performance and volumetric ratio are compared； secondly， the structural design and aerodynamic impacts of different schemes are analyzed according to the classification of wings and head-cones； then， three levels of structural optimization techniques， passive and active thermal protection structural designs are introduced from the perspective of improving the deformation， load-bearing， and thermal protection performance of the morphing structures； finally， we summarize the challenges and problems that are still facing in the work of cross-domain vehicle structures and morphing configuration design， and look forward to the future development direction of related research.

Cross-medium vehicles， capable of operating in both aquatic and aerial environments， have significant potential for applications in ocean monitoring， rescue operations， resource exploration， and environmental protection， making them valuable in a wide range of fields. This paper presents the design of a rotary-wing vehicle with foldable arms， codenamed “Feiyi”， which can operate in three distinct modes： underwater cruising， surface navigation， and aerial flight， with the ability to transition between air and water multiple times. In terms of overall design， “Feiyi” utilizes a rotary body structure to minimize hydrodynamic resistance during underwater motion and to reduce storage space when retracted. The vehicle features a propulsion system with four rotors at the front for flight control and four thrusters at the rear for underwater attitude and movement control. To optimize efficiency in both water and air， “Feiyi” adopts a horizontal posture for underwater navigation and a vertical posture for aerial flight， with attitude adjustments during medium transitions managed by a combination of thrusters. A prototype was developed and tested in various operational modes， including underwater navigation， surface cruising， cross-medium transitions， and aerial flight. The experiments demonstrated that “Feiyi” can achieve a flight altitude of 30 m with a height control precision of ± 0.1 m and an angular control precision of ± 0.5°， representing a significant improvement in control accuracy compared to existing vehicles.

Distributed Unmanned Aerial Vehicle （UAV） swarms can achieve collaborative tracking and dynamic task allocation through autonomous decision-making and information interaction， demonstrating significant application potential in complex and dynamic target tracking control scenarios. To address poor performance of the single swarm in tracking multiple separating moving targets， a self-organized dynamic decision-making method for UAV swarm based on information consensus is proposed. Firstly， considering the communication distance constraints and communication delays between UAVs， a swarm information consensus algorithm based on the latest timestamp forwarding principle is designed to achieve information consensus among UAVs. Secondly， a swarm collaborative decision-making algorithm based on request-response is proposed， realizing member allocation， decision-making and self-organized grouping. Finally， a target tracking control algorithm based on a self-coordination mechanism is designed， achieving adaptive adjustment of control component weights and self-organized target tracking. Compared to existing research on multi-target tracking， the proposed method can realize dynamic target allocation decision-making during the target tracking process and determining the number of UAVs allocated to each target based on the escape risk of each target. Simulation results show that the proposed method can achieve self-organized grouping of the swarm when targets are separating， with a grouping accuracy rate of above 0.9.

Fault diagnosis is one of the key technologies to ensure the safety of liquid rocket engines. The model-based diagnostic methods are limited by the irreconcilable contradiction between diagnostic accuracy and model accuracy， while data-driven diagnostic methods， typified by signal processing techniques， rely heavily on expert domain knowledge. With the rapid development of artificial intelligence and big data， the data-driven intelligent fault diagnosis methods have received extensive attention and achieved great success in a great variety of engineering applications. Therefore， the application modes of the data-driven intelligent fault diagnosis methods in liquid rocket engines was reviewed from the perspectives of model structures and feature engineering of machine learning. The three major challenges faced by the the data-driven intelligent fault diagnosis methods in the practical health monitoring application of liquid rocket engines were further analyzed， and the corresponding solutions based on research achievements of our team were presented， respectively. Finally， review conclusions and future works of the data-driven intelligent fault diagnosis technology were proposed to inspire further exploration in this field.

To improve the safety and public acceptance of logistics drones in complex urban low altitude environments， a real-time dual layer path planning method for urban low-altitude logistics distribution drones is proposed. Firstly， the spatial environment is characterized using grid method and digital elevation model， and a risk perception model for urban low altitude environment based on third-party social risks is established. Secondly， a real-time dual layer path planning model for low altitude logistics drones is proposed. For a single drone in the pre tactical stage， the upper layer model aims to minimize third-party social risk costs and navigation time costs， and an improved A* algorithm is used to plan the optimal expected trajectory. For the coordinated operation of unmanned aerial vehicles in the tactical stage， the lower-level model considers the conflict problem between multiple unmanned aerial vehicles and designs a differentiated conflict resolution strategy based on yaw and hover. With the goal of minimizing the deviation from the optimal expected trajectory cost， a real-time path optimization model for unmanned aerial vehicles is established. Experiments have shown that the upper-level model can reduce operational risk by 15.12% compared to Trajectory Planning considering the Flight Duration（TPFD）， and by 10.61% compared to Trajectory Planning considering the Risk Cost（TPRC）. The lower-level model can effectively generate four-dimensional trajectories with low operational risk， short navigation time， and no conflicts. For a fleet of 50 logistics drones and 100 non-cooperative drones operating in coordination， conflicts can be resolved 60 times within a 10 minutes simulation period， with a flight conflict resolution rate of 100%. The additional risks， flight time， flight distance， and number of grid crossings caused by this can be controlled below 3%.

Addressing the challenges posed by drastic power load variations and high losses in complex transmission systems for distributed electric propulsion aircraft such as electric Vertical Take-Off and Landing/ electric Short Take-Off and Landing（eVTOL/eSTOL）， this paper proposes a high-precision power and energy system modeling and evaluation method applicable to the entire flight process of unmanned aerial vehicles （UAVs）. We first establish an energy system model based on a second-order RC battery model， identifying key parameters through a particle swarm algorithm using battery discharge experimental data， and construct a propulsion system model including propellers， motors， Electronic Speed Controllers （ESCs）， and transmission cables. Subsequently， a working state calculation framework for the UAV power and energy system based on an implicit function equation system is introduced and validated against experimental data. Finally， a simulation assessment of the power and energy system is conducted for a specific project involving a distributed electric propulsion UAV， alongside the optimization design of the transmission cable layout. The results indicate that both the voltage prediction error of the energy system and the errors in the system state calculation framework are smaller than 2%. The model accurately reflects the operational state changes of each component throughout the entire flight process， and the optimized transmission cable layout reduces voltage loss by 17.2% and average power loss by 16.36%， thereby verifying the accuracy and validity of the proposed method.

In response to the challenge of parallel processing of multiple tasks （workpieces） in carrier-based air-craft support operations， which are abstracted as a flexible flow workshop scheduling problem， and the limitations of existing research in the collaborative scheduling of heterogeneous carrier-based aircraft， a dynamic parallel scheduling method that integrates a central scheduling mechanism with a deep reinforcement learning decision model is proposed. Initially， the parallel time series of support operations is equivalently transformed into a serial logical sequence. This transformation ensures compatibility with the flexible flow workshop scheduling model while preserving the characteristic of parallel execution. Subsequently， a Markov model for job scheduling decisions is constructed based on the logical sequences， incorporating the operational differences between manned and unmanned aerial vehicles. Distinct decision models are designed and trained for each type of aircraft. Moreover， a central scheduling mechanism is developed to unify the management of these two decision models， coordinating global positioning， resources， and other situational information. This mechanism disseminates information to the respective decision models to facilitate effective collaboration. Finally， simulation comparison experiments indicate that the proposed algorithm significantly enhances decision real-time performance， even at the cost of marginal scheduling efficiency， compared to optimization methods represented by genetic algorithms. The algorithm effectively balances carrier-based aircraft deployment time and the output time of scheduling methods， making it particularly suitable for rapid deployment tasks in high-real-time and dynamic environments.

In single decision-making models based on reinforcement learning algorithms， the adaptability is often insufficient when handling complex Unmanned Aerial Vehicle（UAV） formation tasks due to limited autonomous decision-making capabilities. To address this， this paper proposes a distributed decision-making method guided by the virtual structure approach integrated with a deep reinforcement learning algorithm. First， to reduce the difficulty of strategy optimization for reinforcement learning algorithms in diverse task environments， the overall task is functionally decomposed. Local task planning is then implemented for individual task scenarios， such as static obstacles， random obstacles， and communication interference. Multiple decision sub-models are constructed along with the design of the calling process between these models. Next， to enhance guidance， the virtual structure method is integrated with the Soft Actor-Critic（SAC） reinforcement learning algorithm to build a distributed decision-making framework. Through decentralized training of each sub-model， the success rate and flexibility of task execution are significantly improved. Finally， a centralized execution approach is adopted， where environmental changes serve as the triggering condition for the dynamic selection and seamless switching betweeen sub-models. This allows the UAV formation to autonomously adjust its formation according to changes in the task environment， achieving the mission objectives while significantly enhancing the overall adaptability and survivability of the swarm. The effectiveness of the method is validated through simulation experiments in multiple scenarios.

As the application of digital twin technology continues to expand in the field of weaponry and equipment， it， as a potentially disruptive key technology in this field， is gradually becoming a core element and an important development means for enhancing the research， development， and design capabilities of aviation weaponry and equipment. Under the development demands of economy， safety， and high efficiency in the aviation field， in-depth exploration of the connotation and key technologies of digital twin technology is crucial for clarifying the future direction of digital simulation technology in the field of aviation vehicles. This paper systematically reviews and analyzes the concept and research status of digital twin， and proposes the key technologies and application development directions of digital twin in the field of aircraft design. The aim is to provide valuable references for the digital construction of the aircraft strength system.

To address the challenges of mutual occlusion， tiny pixels， and complex backgrounds in low-altitude UAV-based object detection， this paper proposes HPRS-YOLO， a small target detection algorithm optimized for UAV platforms. The backbone network incorporates a novel Spatial Pyramid Multi-scale Common Convolution （SPMCC）， which replaces max-pooling-based downsampling with dilated convolution to dynamically adjust the receptive field， thereby enhancing contextual feature extraction. The improved C3K2 module integrates two Metaformer architectures to reinforce structural and textural features of small targets while reducing parameters and maintaining low computational overhead. Additionally， a dynamic upsampling operator， Dysample is introduced to suppress offset overlaps and boundary pixel value confusion， thereby improving target-background contrast. The neck network is redesigned with a Shallow Detail Focus Module （SDFM） to achieve cross-scale feature calibration between terminal layers， emphasizing low-level feature maps to compensate for missing small-target characteristics and preserve spatial integrity of occluded objects. On the dataset VisDrone2019， ablation and comparison experiments are conducted. The results show that mAP_0.5 and mAP_0.5∶0.95 are improved by 5% and 3%， respectively， when compared to the baseline method. Generalization experiments are conducted on the public datasets DOTA， and mAP_0.5 is improved by 2.0%， demonstrating good robustness. Finally， the model is deploying the model on an embedded NVIDIA Jetson AGX Orin device achieves an FPS of 60， demonstrating that HPRS-YOLO guarantees real-time detection capability by optimizing the algorithm design while keeping high accuracy.

The unique advantages of the flying-wing aircraft make it one of the future directions for advanced fighter development. However， in the landing control process， flying-wing aircraft with novel rudder configurations faces such challenges as the nonlinearity， redundancy and coupling of control surfaces， and interference from ship wakes of. To address these issues， this paper establishes a six-degree-of-freedom nonlinear mathematical model of tailless aircraft with multiple control surfaces and proposes a direct force control law based on the Incremental Nonlinear Dynamic Inversion （INDI） control framework. This proposed method designs a direct force trajectory control law， an attitude angle control law， and a speed maintenance control law using the nonlinear incremental dynamic inversion approach. Additionally， a Fixed-Time Disturbance Observer （FTDO） is designed to estimate and compensate for the coupling between the direct force control loop and the attitude control loop， achieving dynamic decoupling between the two loops. Considering various control surface characteristics and the execution capability of thrust vectoring， an integrated direct force control method combining aerodynamic control surfaces and thrust vectoring is designed. The nonlinear control allocation problem is transformed into an incremental linear control allocation problem using the aerodynamic coefficient Jacobian matrix， allowing for rapid online calculation of actual control surface deflection increments. Simulation verification shows that introducing direct force control based on incremental nonlinear dynamic inversion into the landing control law of flying-wing carrier-based aircraft can enhance the ability of such aircraft to quickly correct trajectories and suppress airwake， which ultimately， significantly improves the precision of aircraft landing and provides a solution for the deployment of flying-wing carrier-based aircraft.

Amid the emergence of Industry 4.0 and intelligent manufacturing， significant attention has been garnered， and swift advancement has been undergone by the technology of digital twin. A design framework for digital twins， along with a conceptual modeling approach tailored to this technology， is presented. The proposed digital twin design architecture is anchored in the 3D model， and a comprehensive elaboration is offered of the bidirectional mapping relationship between the target object， pertinent data， and the model， as well as the evolutionary trajectory of the digital twin. Leveraging the object-process methodology， the attributes of objects and processes are extended， and ultimately the minimal general ontology for digital twins is formulated. To illustrate the approach， the UAV electric propulsion system is employed as a case study. Through this lens， a conceptual model of the system is established， and an in-depth account is provided of its composition， behavior， and performance. Promising outcomes are obtained.

Electric Vertical Take-Off and Landing （eVTOL）， a new type of air carrier， has attracted more and more policy support， capital investment and academic attention as one of the actual carriers of low-altitude economy in recent years. The electrical propulsion system is the core power unit of eVTOL， and its performance has a decisive impact on the operational quality of the vehicle. This paper first describes the different architectures of eVTOL electrical propulsion systems， and classifies them to summarize the technical characteristics and requirements of eVTOL electrical propulsion systems. Next， the main domestic and foreign eVTOL machine products and their adopted electrical propulsion systems is compiled and summarized， the performance indexes and technical characteristics of each electrical propulsion system is analyzed， and the current status of the technology level is also clarified. In addition， for the realization of high-performance electrical propulsion system， this paper discusses the research progress of eVTOL motor electromagnetic， cooling， structure， material and electronic control design technology at home and abroad. Finally， based on the aforementioned technical needs and current situation analysis， this paper points out the development trend and technical challenges of eVTOL electrical propulsion system， providing reference for researchers in this field.

Carbon Fiber-Reinforced Polymer （CFRP） composites， known for their exceptional lightweight and high-strength properties， are increasingly valued in the field of aero-engine applications. However， the interfacial issues between fibers and resin matrix significantly limit the full potential of CFRP composites. This paper provides an in-depth discussion on the research progress， existing challenges， and future trends in three key areas： interfacial mechanical properties of carbon fiber/matrix， interfacial damage mechanisms and simulation methods， and interfacial reinforcement mechanisms and strategies. The research indicates that experimental testing， microstructural characterization， and analytical simulation techniques can effectively reveal the influence of interface on the macroscopic mechanical properties of CFRP composites. Interfacial damage exhibits multiple modes including delamination， debonding， and crack propagation， whose evolution processes are governed by the synergistic effects of mechanical loading， thermo-mechanical coupling， and environmental factors. These damage mechanisms can be effectively characterized through numerical methods such as macroscopic finite element simulation， micro-and meso-mechanical models， and multi-scale simulations. Interfacial modification， nano-reinforcement， and novel resin matrix development strategies have significantly enhanced interfacial adhesion properties， thereby improving the mechanical performance of composites. Nevertheless， the application of CFRP composites in the extreme service environments of aero-engines still faces several challenges： limitations in existing interfacial mechanical property characterization techniques， insufficient research on multi-physics and multi-scale coupled damage mechanisms， and inadequate long-term stability of interfacial reinforcement effects. Addressing these critical issues will provide essential theoretical guidance and technical support for enhancing the service reliability of CFRP composites aero-engines and other high-end equipment applications.

To address the challenge of accurately predicting the Remaining Useful Life （RUL） of aircraft engines online， this paper pro-poses a novel RUL prediction method that enhances multi-source sensor temporal feature information. The approach first establishes a prediction network framework by integrating the self-attention mechanism with Bidirectional Long Short-Term Memory （Bi-LSTM） networks. This framework captures the long-term temporal dependencies of multi-source sensor signals and the coupling relationships between their time-varying performance， enabling the extraction of temporal features that in-fluence RUL. To address the potential gradient vanishing issue during training， a residual module is introduced， improving model stability. Additionally， a multi-head self-attention mechanism is employed to extract and enhance key features， leading to dual improvements in both the accuracy and stability of RUL online prediction. Comparative experiments using NASA’s C-MAPSS aircraft engine dataset demonstrate the effectiveness of the proposed method. The results show that the method leverages sensor temporal information to make precise RUL predictions and degradation trend forecasts across a wide range of time and spatial scales. Specifically， the Root Mean Square Error （RMSE） of RUL prediction is reduced by an average of 21.74% compared to other deep learning models， while the coefficient of determination （R²） is improved by an average of 15.81%. This approach offers valuable technical support for the development of aircraft engine health management systems and predictive maintenance strategies.

To address the challenges of limited and incomplete measurable data in fault detection， diagnosis， and prediction of the rudder-loop-system in long-endurance reusable aircraft， this study introduces a digital twin-based approach for health management of the system. First， a digital twin framework for the aircraft’s rudder-loop-system is proposed， incorporating high-fidelity modelling and simulation of mechanical， electrical， control， and dynamic flight load subsystems using AMESim and FLUENT. The integrated model realizes the coupling of three-phase electro-mechanical control with real-time flight dynamic loading. Subsequently， to ensure consistency between the digital twin and the physical system under various operational conditions including normal， rudder surface degradation， and loosening faults， a virtual-physical consistency perception method and a dynamic adjustment mechanism are developed. This enables the digital twin to continuously track physical system changes and maintain synchronization through online fault perception and dynamic updating. Finally， experimental results demonstrate that under both normal and faulty conditions， the digital and physical systems exhibit consistent trends and amplitudes in multiple time-domain indicators of rudder current and deflection angle. Moreover， the expanded data dimensions provided by the digital twin enhance the comprehensiveness of data available for health management. This research highlights the potential of digital twin applications in improving the reliability and maintainability of electric rudders in advanced flight systems.

Flying-wing aerial vehicles have received extensive attention at home and abroad because of their strong stealth performance and excellent aerodynamic characteristics. However， due to the small rotational inertia of the fuselage and the low frequency of the low-order bending modes of the wing， complex aeroelastic problems such as rigid-elastic coupled flutter vibration and aeroelastic vibration are likely to occur in the flight envelope region. An active aeroelastic control method based on estimation of experimental frequency response and robust controller theory is presented for aeroelastic vibration suppression of flying-wing aerial vehicles. The control objective is to simultaneously reduce the aeroelastic vibration of the aircraft structure and enhance the control robustness to external unknown disturbances. An open-loop frequency sweep test in conjunction with the modeling theory of aeroservoelastic dynamics is conducted to estimate the experimental frequency response and obtain the transfer function of a controlled system more closely aligned with the test conditions. Aeroelastic response controllers are designed using robust control theory， and by optimizing the weighting parameters， the number of norms H_{\mathrm{\infty }} of the closed-loop system is minimized to improve the robustness and stability of the system. The effectiveness of the control method can then be estimated by wind-tunnel tests. The numerical results demonstrate that the aeroelastic response controller can significantly reduce the root-mean-square value of the wingtip acceleration of the aircraft in a specific wind speed range up to 35%， which in turn verifies the effectiveness of the robust aeroelastic controller and its robustness.

To address the high-precision landing control problem of carrier-based aircraft under complex environment and strong uncertainty， this paper proposes a direct lift mode active disturbance rejection control method based on offline neural network/online identification. Firstly， referring to the American ‘Magic Carpet’ control system and analyzing its key technical mechanism， the direct lift landing active disturbance rejection control method of carrier-based aircraft is designed. The extended state observer is used to estimate and compensate the total disturbance caused by gust disturbance and system uncertainty. Secondly， according to the evaluation criteria of landing control engineering performance index， the optimal control parameters are selected， and the neural network mapping relationship with flight model uncertainty as input and optimal landing control parameters as output is established. Finally， the active disturbance rejection control parameters are efficiently optimized by online identification. The simulation results show that the proposed method has higher robustness than the baseline controller， and can effectively improve the high-precision landing performance of carrier-based aircraft under interference conditions.

Digital twin technology for aircraft structures reflects the mechanical response and comprehensive performance of an aircraft’s lifecycle through high-fidelity and dynamically updated digital models. To improve the accuracy of structural response predictions， digital twin models often use multi-level， refined simulation techniques. However， this leads to a significant increase in model computation amount and cost， making it challenging to meet the real-time prediction needs for aircraft structural strength during flight. AI-based reduced-order prediction of structural strength is a key technology for real-time prediction of structural responses during flight. By combining high-order digital twin model simulation results with structural loads inferred from actual flight data such as flight parameters and strain， it is possible to rapidly and accurately predict the mechanical response of aircraft structures. This addresses the timeliness issue of aircraft structural performance prediction during actual flights， and has gained increasing attention in the field of aircraft digital twin technology. This paper proposes a method for pixelating structural cloud maps and load inputs to express spatial relationships between point cloud data. To construct an intelligent prediction method for mechanical responses Based on load inputs， a convolutional neural network with cross-layer connection mechanisms is also introduced. Results from numerical experiments on the wing structure under 329 conditions show that the pixelation method can retain structural response characteristics， while making the data compatible with pixel-based data processing methods like convolution. Compared to traditional convolutional neural networks， the proposed residual neural network model achieves improved prediction accuracy and reduced loss. Additionally， this intelligent method achieves more than two orders of magnitude efficiency improvement compared to traditional simulations. The predicted stress distribution shows a high similarity to finite element simulation stress distributions， highlighting its application value in the digital twin of aircraft structures.