A pose graph optimization method employing nonlinear factor recovery was created in order to accomplish high-precision vehicle pose estimation in urban road situations. This method successfully incorporates vision, inertial information, and global navigation satellite system (GNSS) data inside the factor graph framework. To address the issue of unclear covariance estimation in previous pose graph optimization algorithms, the nonlinear factor recovery algorithm extracts the required information from the dense prior factors generated by marginalization, replacing the prior factors with relative pose factors and conducting optimal estimates for the covariance matrices of these factors. A procedure utilizing visual-inertial odometry for the active detection of GNSS signal anomalies has been designed, capable of adaptively adjusting the covariance matrix of GNSS signals. Throughout the pose graph optimization process, these strategies guaranteed consistency of factor information. Tests conducted on datasets and in real road environments indicate that this approach significantly improves the fusion efficiency of multisource asynchronous signals, providing robust and globally consistent high-precision pose estimation results.

Fault detection technology is an important technology to ensure the safe and stable operation of the electric-hydraulic servo actuator (EHSA). Owing to the electro-hydraulic servo mechanism’s issues with unequal detection time intervals and limited detection times, utilizing such detection data directly in conjunction with the current algorithm to identify EHSA failure may result in decreased detection accuracy and model performance. In order to solve this kind of problem, a fault detection method of an electro-hydraulic servo mechanism based on non-equal interval data was proposed. First, the historical detection data of the equipment was isolated by combining the historical detection data of electro-hydraulic servo mechanisms of various devices in the same batch using the M-H algorithm. Second, the contribution graph approach was utilized to examine the fault-related indicators after statistics were constructed using the kernel principal components analysis (KPCA) method, which analyzed the discrepancies between principal components and served as the basis for fault identification. When the number is limited and the data are not evenly spaced, the suggested approach has a high defect identification rate. The effectiveness of the method is verified by the historical detection data of a certain type of electro-hydraulic servo system.

The Wilson-ρ_∞ technique is formed by establishing a relationship between ρ_∞ and θ, as the high-frequency dissipation of the Wilson-θ method cannot be properly controlled by ρ_∞ (spectral radius at infinite frequency). The numerical performances of this approach are compared with those of the Generalized-α method. In the Wilson-ρ_∞ method, there are two different θ for a given ρ_∞. The characteristic roots of the Jacobi matrix corresponding to both θ are different, and the corresponding Wilson-ρ_∞ method has different numerical performances. A better θ is recommended according to the properties of the spectral radius. In addition, an analog system of a single degree-of-freedom forced vibration system is constructed with the dissipation and frequency of the Wilson-ρ_∞ method, and the initial conditions of which and the forces acting on the analog system are the same with those of the original system. It is evident that the steady state responses have no cumulative amplitude errors and phase errors, and the results of the Wilson-ρ_∞ method match the analytical solutions of the analog system.

The multi-indenture characteristics of civil aircraft components and the multi-echelon maintenance levels in the actual repair sites are the key factors in the analysis of civil aircraft maintenance support. Therefore, level of repair analysis is an important component in carrying out civil aircraft operation support activities. Existing economic models for level of repair analysis have the problem of repeated accumulation of costs between upper-level parent components and their subordinate child components when making decisions on discarded or moved items. In this paper, the constraint relationship between parent and child parts is studied, and the total cost data preprocessing, matrix is introduced to solve the problem of repeated accumulation of costs of parent and child parts. Based on the cost data preprocessing matrix the constraints are simplified and improved, and the exact algorithm of the LINGO18.0 software is used to model and solve the problem. The findings demonstrate that, in comparison to heuristics and other approximate algorithms, the model and its solution algorithms suggested in this paper can support decision-making for the development of the maintenance design and support of the civil aircraft. Whether it is a two-indenture and two-echelon model or a three-indenture and three-echelon model, the global optimal solution can be obtained in a shorter amount of time, and the maintenance decision-making cost can be reduced by 37.9% and 27.8% separately.

Addressing the issue that the combination of emojis and text data may alter the original semantics, and the mechanism of their interaction with text data has not been fully explored. For this reason, textual sentiment classification incorporating dual emoji attention mechanisms is proposed in the paper. First, a BERT pre-training model is used to obtain the dynamic word vector representation of text; then a CNN-BiGRU dual-channel model is constructed to extract local and global features respectively; after that, the Emoji2vec model is used to obtain the emoji vector representation and construct a dual emoji attention mechanism, which strengthens the key information of the combination of emoji and text from the level of local and global emoji attention mechanisms respectively; then the output feature vectors are fused to classify emotions. In order to verify the effectiveness of the proposed model, contrast and ablation experiments were set up. The Emoji-phone and EmojifyData datasets were used for sentiment classification training, and the findings indicate that the model in this article outperforms the more recent RoBERTa-3xBiGRU model by 0.0176 and 0.0166, respectively.

This paper proposes a cross-layer efficient phase-aware Transformer segmentation algorithm for colorectal polyps in order to address the issues of irregular shape of the lesion region, fuzzy edge contour, and high similarity with normal region, which result in the loss of detail information and mis-segmentation of the lesion region. Firstly, the pyramid vision Transformer encoder is used to extract the global semantic information and spatial details of the input feature map layer by layer, and to analyze the colorectal polyp lesion features at multiple scales; secondly, the polarized self-attention module is used to regressively predict the lesion features, and to deepen the correlation of the semantic information of the features;thirdly, the high-efficiency phase-aware module is designed to extract the global and local information to precisely. The final one is the cross-layer fusion and propagation module, which enhances the rate of advanced feature reuse by integrating the edge details. Experiments were conducted on five datasets: CVC-ClinicDB, Kvasir-SEG, ETIS-LaribPolypDB, CVC-ConlonDB, and CVC-T, achieving Dice coefficients of 0.940, 0.923, 0.801, 0.810, and 0.896, respectively. This demonstrates superior segmentation performance over existing networks such as CaraNet and MSRAFormer. Both colorectal polyp images with fuzzy edges and complicated spatial organization exhibit great segmentation accuracy, according to the evaluation results.

In view of the problem that the electrical connector is subjected to step stress random vibration during operation, which leads to fretting wear of the electrical connector and the reduction of contact performance, the step stress random vibration test is carried out. The electrical capacitance tomography (ECT) is used to detect the characteristic value of wear debris between the electrical connector contacts in the process of fretting wear. Contact resistivity and the characteristic value of wear debris are used to study the wear degree and degradation law of contact performance under step stress random vibration conditions. The maximal information coefficient (MIC), which has good robustness and can characterize the nonlinear relationship, is introduced to analyze the correlation between the characteristic value of wear debris and contact resistance, and the dimension is reduced through mic screening to improve the prediction accuracy of the model. The findings demonstrate that under the random vibration of step stress, there is a stepped change trend in the contact resistance, the total characteristic values of wear debris, and the characteristic values of wear debris. Through the calculation of the maximum information coefficient, it is found that the total amount of wear debris characteristics is strongly correlated with the contact resistance. The results of energy spectrum analysis are consistent with the test results. The average absolute error of CHIO-Elman neural network performance degradation model optimized by MIC screening is less than 4%.

Addressing the critical challenges of data class imbalance, low detection accuracy, and insufficient real-time performance in existing intrusion detection systems for communication-based train control (CBTC) systems, this study proposes a novel lightweight intrusion detection method for CBTC onboard equipment based on a fused Naive Bayes-extreme gradient boosting (NB-XGB) model. The proposed approach operates through two coordinated phases: an offline training phase and an online detection phase. In order to identify an ideal feature subset with greatest relevance and minimal redundancy, the CICIDS2017 and CBTCset datasets are preprocessed and then subjected to correlation-based feature selection during the offline phase. The extracted features are then balanced using a hybrid sampling approach that combines edited nearest neighbors (ENN) undersampling and synthetic minority over-sampling technique (SMOTE), after which the naive Bayes (NB) and XGBoost models are integrated through static weighted fusion to establish the classification foundation. To provide effective real-time detection for online deployment, knowledge distillation technology transfers the learned information from the trained NB-XGB fusion model to a lightweight multilayer perceptron (MLP) student model. The method’s efficacy is demonstrated by experimental evaluation on the CICIDS2017 and CBTCset datasets, where the NB-XGB fusion model outperforms comparative models such as K-nearest neighbors (KNN), NB, and isolation forest (IForest) with notable accuracies of 0.9961 and 0.9557. It simultaneously demonstrates the effectiveness of the proposed method. Additional lightweight validation confirms the distilled model’s inference speed of 0.16 ms per sample, sufficiently verifying the solution’s real-time capabilities for practical CBTC intrusion detection scenarios.

The drive architecture with many power converters in parallel is typically used in distributed electric propulsion systems, which are a key avenue in the development of electrification for all-electric aircraft, electric ships, and other transport vehicles. With the increase of the number of converters, electromagnetic interference will be intensified, which directly threatens the safe and stable operation of the system. In order to suppress the conducted interference generated by parallel drive systems, a modulated wave phase-shifting control strategy is proposed. Based on the in-depth analysis of the coupling mechanism of electromagnetic interference noise in the system, the conducted interference equivalent model of the parallel drive system is established through the cascade of a multi-port network model. By introducing real-time position current compensation, the modulation wave of each driver is controlled by phase-shifting, so that the common mode noise generated by each driver cancels out each other. The simulation and experimental verification are carried out in a MATLAB/Simulink simulation environment and a darkroom experiment environment for a dual-motor parallel drive architecture. The results show that when the phase difference of the driver modulation wave is 360°/m(m is the number of parallel drivers), the noise reduction is most significant. This technique greatly lowers the system's conducted interference without appreciably altering the system performance index, and it works well with parallel drive systems that have the same drive module specifications. It has a certain reference value for the design of a distributed electric drive system.

The nonfragile asynchronous control method is proposed for Takagi-Sugeno (T-S) fuzzy Markov jump systems (MJSs) with hybrid cyber-attacks and time-varying delays. Firstly, the Markov jump system model is established by utilizing the T-S fuzzy method under fuzzy rules. The hybrid cyber-attacks consisting of deception attacks and denial-of-service (DoS) attacks are considered in the communication network, and the dynamic event-triggered communication mechanism is adopted in the measurement channel to reduce unnecessary data transmission. Secondly, sufficient conditions for the stochastic stability of the closed-loop system are derived in terms of the Lyapunov-Krasovskii method and linear matrix inequality technique. The linear matrix inequality technique is used to solve the nonfragile asynchronous control gain matrices. Lastly, simulation examples verify the correctness and effectiveness of the proposed method.

To improve the resilience of the airport infrastructure network under rainstorm disaster, the airport infrastructure network topology model is established by considering the functional characteristics of airport infrastructure, and the complex network characteristics of airport infrastructure are analyzed. The service efficiency function of the airport network model is constructed by introducing the number of flights, passenger capacity and route distance between the airport nodes. The relevance index of the airport nodes is determined by comparing the network model service effectiveness before and after the airport node failure in order to identify the important airport nodes in the network. The resilience triangle theory serves as the foundation for the resilience recovery model of the airport infrastructure network, which is designed to investigate the best practices and recovery order for airport nodes in the event of failure for major airport nodes, major regional airport nodes, and multi-regional airport nodes during rainstorm disasters. The results show that the established model is presented as a small-world network, which is characterized by high agglomeration. The five key airport nodes that have a greater impact on the network service efficiency are Guangzhou Baiyun Airport, Beijing Capital Airport, Shenzhen Baoan Airport, Hangzhou Xiaoshan Airport and Shanghai Hongqiao Airport. Compared with the recovery strategy based on importance degree and node degree, the recovery strategy based on network resilience has the best recovery effect, such as the network resilience value based on node degree and importance degree under the key airport nodes failure is 0.889, 0.907, while the resilience value based on the network resilience recovery strategy is 0.915. Compared with the airport nodes in North China and Southwest China, the failure of the airport nodes in East China has a greater impact on the operation efficiency of airport infrastructure network.

The advanced fighter aircraft's onboard electronic equipment is developing towards high power, high integration, and miniaturization, which results in a continuous increase in heat generation and heat flux and poses a serious challenge to airborne thermal management. With a transport distance of 2.30 meters and ammonia as the working fluid, a high-power dual compensation chamber loop heat pipe（LHP） was created in response to the aforementioned specifications. A comprehensive and systematic experimental study on the dual compensation chamber loop heat pipe was conducted, mainly focusing on its startup characteristics, heat transfer capacity, and thermal resistance change. With a heat transfer capacity of over 900 W, the testing findings demonstrate that the dual compensation chamber loop heat pipe can successfully accomplish startup and run smoothly under a variety of evaporator attitudes, including horizontal, favorable, and unfavorable attitudes. The system thermal resistance of the dual compensation chamber loop heat pipe first rapidly decreases and then gradually increases with the increase of the heat load, and the minimum value is about 0.063 ℃/W. This work provides a new technological means and a feasible solution for efficient thermal management of future airborne systems.

As a core propulsion device for spacecraft, the liquid rocket engine relies heavily on its electro-mechanical actuation system (EMAS) for precise thrust regulation and control. Permanent magnet synchronous motors (PMSM) are widely applied in EMAS. However, traditional proportional-integral (PI) current control exhibits inherent limitations when confronted with variations in motor electrical parameters. Accordingly, research on model predictive control (MPC) for PMSM is conducted in this paper.The mathematical models of PMSM in different coordinate systems are established, and predictive speed and current control strategies are designed based on the fundamental principles of MPC, including the corresponding prediction models and cost functions. A simulation model is built on the MATLAB/Simulink platform to compare the performance of MPC with field-oriented control (FOC). Simulation results show that compared with FOC, MPC reduces the steady-state error by approximately 99%, eliminates overshoot, and shortens the response time by about 80% in position step response. It delivers superior dynamic tracking capability under swept-frequency input and cuts the speed fluctuation amplitude by around 80% under variable load conditions. Experimental verification is carried out on a motor back-to-back test platform. The test results prove that MPC achieves smoother response and smaller oscillation when the input signal changes, and possesses stronger adaptability to large load disturbances. Applied to PMSM control in the EMAS of liquid rocket engines, MPC delivers better control performance, effectively improves the overall control effect of the EMAS, and provides an optimized solution for relevant engineering applications.

The DINO algorithm, tailored for remote sensing image detection, has garnered significant traction in recent research circles. Remote sensing object detection mandates algorithms to excel in both fine-grained feature extraction and large-scale spatial search capabilities. However, an improved DINO algorithm is proposed, the inherent multi-layer encoder-decoder architecture of the vanilla DINO algorithm incurs substantial spatial and computational overhead, posing notable impediments to real-time inference performance. To address these limitations, this study capitalizes on the inherent suitability of parallel large-kernel convolution structures for remote sensing image processing, proposing a single-layer efficient hybrid encoder architecture to enhance the parameter efficiency of the DINO algorithm framework. Within this novel encoder structure, we redesign the core module based on high-efficiency convolution operations and integrate structural parameterization techniques. This design strategically reduces both the number of trainable parameters and floating-point operations during inference, thereby effectively mitigating the latency bottleneck of the original DINO algorithm. Experimental evaluations on the NWPU VHR and DOTA datasets demonstrate that the improved DINO algorithm achieves marginal yet consistent performance gains, with mean average precision (mAP) improvements of 1.8% and 3.8% respectively compared to the baseline. Most notably, the proposed modifications yield substantial reductions in model size and computational complexity. When benchmarked against state-of-the-art remote sensing detection algorithms, the improved DINO algorithm maintains competitive detection accuracy while outperforming counterparts in terms of parameter efficiency, computational cost, and inference speed.

In the unknown environment without prior conditions, the path planning and navigation of an underwater autonomous underwater vehicle (AUV) is a big challenge. This study presents a perceptive navigation approach without a prior geomagnetic map. It achieves efficient path planning and geomagnetic map creation by combining deep reinforcement learning (DRL) with a simulated annealing (SA) algorithm. A deep Q network (DQN) is built to explore the environment of the carrier, collect local geomagnetic data, and then use the collected data to train a regression model to predict the global geomagnetic map. At the same time, the simulated annealing algorithm is used to optimize the path of an underwater AUV to avoid the local minimum problem of carrier space search. The success of the suggested approach is confirmed through a number of simulated tests in terms of path length, exploration efficiency, and geomagnetic map correctness. The results show that this method can significantly improve the navigation performance of underwater AUV in an unknown environment, and provide a new way for the construction of geomagnetic maps.

An insulin pump is an advanced device used for intensive insulin therapy in diabetic patients. Failure of the insulin pump sets can disrupt normal insulin delivery, leading to abnormal blood glucose elevations and potentially causing diabetic ketoacidosis, which can be life-threatening. Establishing a mathematical model to describe the fault mechanisms of insulin pump sets is fundamental for its fault diagnosis. Insulin pumps, however, provide difficulties for modeling and fault mechanism analysis due to the stiff and elastic restrictions of the needles and tubes, as well as the multi-domain interactions between the fluid (insulin) and the solid parts of the pump. In response to these challenges, this paper establishes a mathematical model of fluid transmission in insulin pumps under both healthy and faulty conditions, based on power flow theory, focusing on two typical faults: blockages and leaks. The impact of these faults on fluid flow within the insulin pump sets is quantitatively analyzed. The computational results of the proposed model showed a maximum error of 0.57% when compared with professional fluid dynamics software simulations. Furthermore, the impact of varying degrees of leakage and blockage faults on the output flow rate and pressure of the insulin pump system is analyzed. Specifically, for blockage faults, it was observed that when the blockage layer thickness is less than 0.6 mm, the changes in insulin output flow rate and the pressure within the insulin pump chamber are not significant. However, the blockage has a considerable impact on both flow rate and pressure when the thickness of the blockage layer surpasses 0.6 mm. The impact grows with the thickness of the blockage layer.

The vibration parameter-based bridge inspection method was proposed under the coupling effect of aircraft and bridge, aiming at the health bearing performance of taxiway bridge under the compound effect of earthquake, aircraft load, and damage grading coupling field. The aircraft-taxiway bridge coupling vibration damage model was set up through the finite element simulation and on-site experiment; the model parameters were adjusted according to the measured data, and the error of the model was controlled within 7%. Input the adjusted seismic wave, simulate the mechanical properties of the taxiway bridge under different aircraft load, damage degree and seismic load, establish the vibration performance analysis model of taxiway bridge under multi-field coupling, and realize the assessment of the mechanical properties of the bridge's healthy bearing. Two working conditions were selected to verify the feasibility of the model. The aforementioned study findings will offer a crucial theoretical foundation and point of reference for the identification and assessment of taxiway bridges using vibration characteristics.

To address the issue of individual variability in workload tolerance, this study establishes a quantitative model for assessing controller workload. A test was devised for the purpose of collecting data from 24 area controllers both before and after their workday. Based on the test data, variables were selected for analysis that were deemed to be sensitive in terms of describing the individual load. Three dimensions were included in the comprehensive evaluation index system: cognitive workload, physiological reaction load, and psychological perception load. A model for the individual load index of controllers was developed. The optimal weights of the individual load index were determined through the application of the entropy weight-CRITIC combination method. The individual workload index of each controller was subsequently calculated. It was determined that there are notable discrepancies in the individual workload of controllers. The cognitive workload index is a principal indicator of the magnitude of the individual workload index for controllers. The cognitive workload is a principal index for determining the magnitude of the controller’s individual workload index. Additionally, there is a positive correlation between cognitive workload and the controller’s capacity for information retrieval, decision-making, and reaction. To further investigate the factors influencing the growth of controllers’ cognitive workload, the reaction time, gaze time, and sweep time of controllers under five distinct traffic levels were quantified. Additionally, pre-post and post-post paired tests were conducted, and one-way analysis of variance was performed between groups with the same indicator. The results showed that while controllers’ reaction ability was more affected by flow parameters, their decision-making and searching abilities were more vulnerable to cumulative load.

There is a high correlation between an object's photometric data and its shape, size, material, and movement. In order to classify the type of satellite platform using the photometric data, a deep neural network based satellite platform classification method is proposed. Preprocessing techniques like as interpolation, smooth filtering, distance correction, and phase correction are applied to the photometric data needed for network training. Convolutional long short-term neural network is constructed to extract the spatial and temporal features of space objects from photometric data, and the average classification accuracy of rocket, satellite Iridium, satellite GlobalStar and space debris is 73.8%, better than the 70.43% accuracy of the convolution neural network. Additionally, the deep neural network is further tested using simulated photometric data from seven satellite platforms, including White Cloud (WC), defense meteorological satellite program (DMSP), future imagery architecture (FIA), and geosynchronous space situational awareness program (GSSAP). The classification accuracy of the satellite platforms is significantly increased to better than 90%.

Objective Optical solar reflector (OSR) exhibit a very low absorptance-emittance ratio, ensuring stability and high performance in space applications. They efficiently aid in cooling or heat dissipation when attached to the exterior of radiative cooling panels. Understanding the impact of space environment on the OSR and the degradation mechanisms of their thermal control properties is crucial for designing long-lasting thermal control systems. Through analysis of temperature data from cameras onboard four satellites operating in geostationary orbit for 7 years, 4 years, 3 years, and 2.5 years, coupled with thermal simulation analysis models, the degradation model of the OSR solar absorptance solar absorptance is obtained. The deterioration model of the OSR solar absorptance solar absorptance is derived by analyzing temperature data from cameras on four satellites that have been in geostationary orbit for seven, four, three, and two and a half years. This data is combined with thermal simulation analysis models. The results reveal that the solar absorptance of the OSR were 0.125 in the first year of operation, degraded to 0.134 after 2 years in orbit, and changed to 0.175 after 7 years, with the degradation curve exhibiting linear characteristics and an annual degradation rate of approximately 0.74%. Additionally, based on this degradation model, the solar absorptance is predicted to be 0.195 after 10 years. OSR has strong spatial adaptability and stability and the results of this article offering guidance for subsequent in-orbit temperature control and new thermal control system designs.

The uncertainty of the model should be statistically assessed as the foundation for model selection, as there are errors between the digital twin model and the actual system that must be reduced. However, the satellite digital twin model exhibits multi-dynamic, multi-spatial scale, and multi-physical field coupling properties. Additionally, the numerical solution will reveal the stiffness problem of ordinary differential equations and the multi-scale problem of partial differential equations. If many telemetry parameters are updated at the system level, the results will not converge. A multi-granularity and negotiation model updating framework for satellite digital twin method was proposed. The parameters were grouped by correlation analysis and frequency domain analysis. Multi-granularity digital twin models were built based on the requirements, and various granularity models of satellite subsystems and components were produced. The coupling relationship between the satellite structure of different levels was studied, and a negotiation updating method was proposed. Real on-orbit telemetry data were used to verify the framework. According to the research findings, the proposed method updating approach outperforms the unused one by over 50% in terms of accuracy, and the updating results are more thorough and methodical.

The structural safety factor of an aircraft, defined as the ratio of design load to service load, is a key parameter in aircraft design. Traditional design methods rely heavily on engineering experience, leading to subjective safety factor values and insufficient objectivity in quantifying uncertainties. For advanced aircraft requiring refined design, the uniform safety factor applied across all components results in overly conservative designs that limit ultimate flight performance. In order to solve this limitation, it is necessary to develop a sub-regional differentiated safety factor design method to better explore the material properties and design space on the premise of ensuring the reliability design requirements. In this paper, probabilistic reliability design optimization theory is used to study the uncertainty of the structural system, and the mapping relationship between structural reliability and sub-regional differentiated safety factors is established, and the design method of sub-regional differentiated safety factors is developed. Using the simplified engineering model of the wing-tip structure as an example, it is demonstrated that, assuming the design requirement of 99% structural strength reliability is met, the sub-regional differentiated safety factors in the majority of the structure's subregions are less than the unified safety factor of 1.45, resulting in a relatively lighter design weight of up to 3.926 kg.

As a mature near-space aerial platform, high-altitude scientific balloons exhibit unique advantages in astronomical observations. However, the conventional single-axis attitude pointing control method for balloon-borne gondola platforms, which ignores the roll effect, can no longer fully meet the requirements of emerging space science application scenarios. Therefore, in-depth research on the motion and control characteristics of the pitch/roll channels of the balloon-gondola system is required to improve the overall performance of the attitude control system. Accordingly, the dynamic characteristics of the pitch/roll channels of the balloon-gondola system are modeled using the Lagrangian equation method, and an accurate modal calculation approach is proposed. After linearizing the established dynamic model, its controllability and observability are analyzed, and two control strategies, namely torque damping and active compensation, are put forward. For the torque damping control strategy, a linear quadratic regulator (LQR)-based controller and a Kalman observer are designed, and verification is carried out via Simulink simulation. The proposed dynamic modeling and modal calculation methods for the pitch/roll channels of the balloon-gondola system further reveal the motion characteristics of the coupled system. The proposed control strategies and corresponding simulation results provide an important reference for the design and optimization of attitude control systems for high-altitude scientific balloon gondolas.

An improved proportional-derivative (PD) controller with adaptive capability is designed for the rigid-elastic coupled launch vehicle, aiming to achieve high-quality tracking of guidance instructions. The controller introduces an adaptive augmenting control (AAC) module, enhancing its parameter adjustment capabilities so that it can adaptively adjust both proportional and derivative coefficients simultaneously. Furthermore, an adaptive filter module is incorporated, with amplitude reduction complementing phase lag in the amplitude-phase characteristics of the transfer function. This filter adjusts its filtering effect adaptively based on the identification results of the elastic frequency. When applied to the pitch angle rate feedback channel, this filter effectively stabilizes the phase angle of the first-order elastic mode and the amplitude of the higher-order elastic mode. Numerical simulation results demonstrate that the designed adaptive improved PD controller not only has a simple structure and is easy to select parameters, but also exhibits excellent adaptive capabilities and significant elastic vibration suppression effects. Its remarkable engineering application value is demonstrated by its suitability for time-varying systems and its capacity to preserve system stability even in the presence of large model parameter variations.

The effectiveness of flight operations and the quality of airport services are directly impacted by airport gate assignment options. In real-world operations, unexpected events such as airfield accidents may lead to the temporary closure of local gates and an infeasible assignment plan. It is urgent to implement gate emergency reassignment under resource constraints. This paper proposes an Epsilon constraint-based column generation optimization algorithm for this problem. In particular, we develop a bi-objective optimization model based on set partitioning with the goal of minimizing assignment plan deviation and increasing solution efficiency. Then, an Epsilon constraint-based column generation optimization algorithm is designed to efficiently obtain high-quality solutions. Numerical experiments are conducted based on real-world operational data from an international airport. The results demonstrate that the proposed method performs well on main metrics such as the bridge boarding rate and flights adjustment efficiency. In particular, the cross-region adjustment proportion of our solution is 52.34%, which is significantly lower than the comparison methods, and effectively improves the airport operational efficiency in emergency scenarios.

Gravity-aided inertial navigation is an important technology for enabling long-term autonomous navigation of underwater vehicles. However, the gravity anomaly observation is coupled with inertial navigation state parameters in a complex manner, and improper selection of estimated parameters may lead to reduced filtering accuracy or even filter divergence. To clarify the real-time estimation effectiveness of different navigation state parameters, this paper conducts modeling, analysis, and verification of the observabal degree of state parameters in a gravity-aided inertial navigation system. First, an error-state filtering model for gravity-aided inertial navigation is established. A 13-dimensional state vector is selected, including attitude errors, velocity errors, position errors, gyroscope biases, and accelerometer biases. The gravity anomaly observation equation and its partial derivatives with respect to each state parameter are derived, thereby clarifying the relationship between the observability matrix and the estimability of navigation parameters. Second, observabal degree analysis models based on the covariance matrix, observability matrix, and Lie derivatives are constructed, respectively, and the applicability and consistency of different methods are compared. Based on a simulated underwater vehicle trajectory in a certain sea area, the variation patterns of the observabal degree of different state parameters with navigation time, gravity field characteristics, and maneuvering conditions are analyzed. The experimental results show that longitude, latitude, eastward velocity, and northward velocity have relatively high observabal degrees and constitute a preferable combination of estimated states for gravity-aided inertial navigation. In contrast, gyroscope and accelerometer biases have relatively low observabal degrees and are not suitable to be directly used as the main feedback correction parameters. On this basis, fixed parameter combinations and a dynamic parameter combination based on an observabal degree threshold are further designed for filtering verification. The results indicate that joint estimation of position and horizontal velocity improves navigation positioning accuracy by 42%, while dynamically adjusting the state parameter combination further improves positioning accuracy by 5%. The research results provide a basis for filtering-state selection, equation-structure design, and feedback correction strategies under maneuvering conditions in gravity-aided inertial navigation.

The need for data security and privacy protection makes it difficult to directly share some flight operation data. Federated learning (FL) achieves data availability that is invisible, but when faced with honest but curious opponents, it still faces the threat of reverse attacks and the risk of privacy leakage. The inadequate performance of FL in striking a balance between privacy protection and model utility is addressed by the suggested artificial bee colony and dual Rényi differential privacy federated learning (ABC-2RDP-FL) flight operation data privacy protection algorithm, which is based on dual Rényi differential privacy (RDP) and artificial bee colony (ABC). In ABC-2RDP-FL, a dual RDP protection mechanism is designed to measure privacy budgets more strictly and improve privacy protection performance. After that, an ABC-based FL hyperparameter optimization approach is suggested to enhance model performance while striking a compromise between model utility and privacy protection. Finally, the effectiveness of the proposed method was validated using public data and flight operation data.

Sentiment dictionaries are essential for sentiment analysis in tourism reviews since they provide valuable prior knowledge for identifying lexical emotions. The selection criteria for seed word sets in the conventional method of building the domain sentiment dictionary typically only utilize semantic vectors or term frequency statistics, which leaves the seed word set with insufficient emotional representation and, consequently, impairs the vocabulary emotion recognition accuracy. Therefore, we propose a method of sentiment seed word set selection based on a multiple feature fusion strategy and emoji integration. This method integrates corpus statistical features, emotional intensity features, and lexical semantic features as the screening criteria for various emotional seed word sets, ensuring a high match between seed vocabulary and corpus characteristics, improving the representativeness and coverage of the seed word set effectively. At the same time, emoticons are introduced to enhance the emotional expressive capabilities of the seed set and improve the accuracy of emotional classification of vocabulary by adding emotional aspects that emotional vocabulary might overlook. Finally, it constructed a fine-grained sentiment dictionary for the tourism field. According to experiments, sentiment analysis employing the tourism field’s sentiment dictionary enhances the accuracy rate by 0.0949, recall rate by 0.0896, and F₁ value by 0.0923 on average when compared to other Chinese general dictionaries in the tourism corpus.

During close-range flying tasks, helicopters are frequently struck electrical lines. Helicopter terrain awareness and warning system (HTAWS) is an essential instrument for preventing helicopter crashes, however it is challenging to reliably assure helicopter flight safety in the absence of a power line database. This research suggests a machine learning-based approach for building a power line database using satellite imagery. Aiming at the problem that YOLOv5 is insensitive to small target detection and has a high rate of missed detection, an improved YOLOv5 is proposed to identify the pylons in satellite images. Than the geospatial data abstraction library (GDAL) module was then used to calculate the longitude and latitude of the pylons. A pylons height acquisition method based on the shadow of pylons was proposed. The prediction of pylon connections was investigated using group features of pylons and graph neural networks. The simulation results demonstrate that the approach presented in this research can determine the longitude and latitude of pylons with 94.65% accuracy when determining the tower’s height. The Matthews correlation coefficient (MCC) value predicted for the connection of pylons is 0.479, which can establish a power line database that meets the requirements of helicopter power line collision warning.

The uniformity of the flow field inside a large composite material curing oven directly affects the curing quality of thermosetting resin composite parts. To improve the curing effect within the curing oven, this paper establishes a model of a large composite material curing oven including cured parts, based on computational fluid dynamics (CFD) theory. The simulation study focuses on how the curing oven’s internal flow field is affected by the air supply method, the number of air supply inlets, the number of cured pieces, and their spatial layout within the oven. The results indicate that the top supply and bottom return air supply method generally outperforms the side supply and side return air supply method in terms of velocity field uniformity and temperature field uniformity. When the number of air supply inlets is 5, the velocity field and temperature field within the curing oven exhibit better performance. As the number of cured parts increases, the air velocity and temperature inside the curing oven decrease, and the uniformity of the velocity field and temperature field will weaken. The temperature field and velocity field inside the curing oven operate exceptionally well when the cured pieces are positioned in a "triangular" pattern. The experimental results show that the maximum temperature error between simulation data and experimental data is 4.2 ℃, which is within a reasonable range. This proves that the simulation results are reliable and can provide a reference for design work.

In order to address the issue that traditional radar signal recognition techniques have trouble successfully identifying unknown modulated signals in practical situations, this paper suggests a multi-level radar signal open-set identification technique based on K-means clustering and Support Vector Machine (SVM) pre-training. After performing the multisynchro squeezing transform (MSST) on radar signals, the discrete wavelet transform (DWT) is employed to extract features from the preprocessed time-frequency images. An SVM classifier is trained using known radar signal data during the training phase. The classifier is used to distinguish between known and unknown modulation types during the testing phase in order to achieve open-set radar signal identification. Subsequently, K-means cluster analysis is applied to unknown radar signals, further classifying unknown modulation modes into different clusters, thereby expanding the recognition scope of radar signal modulation types. Experimental results demonstrate that the proposed method achieves a recognition accuracy of over 90% for both known and unknown signals at a signal-to-noise ratio (SNR) of −4 dB, effectively recognizing unknown modulation types.

The Transformer-based rolling bearing fault diagnosis algorithms have a quadratic increase in computational complexity with the input time window, leading to a decrease in the real-time performance of the model inference. To address this problem, a lightweight Transformer fault diagnosis model based on improved linear attention is proposed. Firstly, the improved linear attention is proposed to reduce the quadratic computational complexity, which uses the strategy of changing the computation order of the dot product. Secondly, by substituting the suggested feature bias mapping function for the Softmax global mapping function, the enhanced linear attention feature recovery block is suggested, which reduces the computational burden of utilizing the global acceptance field. At the same time, the bias function has an efficient feature focusing mechanism, which demonstrates significant anti-noise interference properties by strengthening the connection between similar features and weakening the coupling between dissimilar features. Then, the feature diversity recovery block is used to approximate the performance of the original self-attention after global activation and to recover the modeling ability for long-term temporal dependencies. Experiments are conducted on three mechanical failure datasets from Xi'an Jiaotong University and the University of Ottawa. Compared with seven typical models, namely CLFormer, ConvFormer-NSE, MCSwin-T, MobileNet, MobileNet-V2, ResNet18 and MK-ResCNN, the results show that the proposed model outperforms the above models in terms of accuracy and real time performance, and has good robustness in heavy noise environments at the same time. To create a comprehensible connection between the suggested approach and the prediction outcomes, visualize the feature bias mapping function's weight information. Finally, the effectiveness of the proposed modules (feature bias mapping function, feature diversity recovery module) is verified by ablation experiments.

With the rapid development of the aviation industry, the safe flight of aircraft has become particularly important. A flight-level anomaly identification method based on unstable approach events during aircraft approach is presented to identify anomalous events in the aviation field. The method, called PVAE-WGAN, combines variational auto-encoders (VAE) and Wasserstein generative adversarial networks (WGAN), and uses a Pareto distribution to simulate the probability distribution of anomalous cases. The generators of VAE and WGAN are shared, and the hidden variables randomly sampled from the normal distribution and Pareto distribution are taken as input to the generator. The reconstructed output samples are taken as positive and negative samples, respectively. The Wasserstein distance is used as a measure between the distribution of positive and negative samples fitted by the model and the true distribution, so that both the generator and the discriminator gain the ability to distinguish anomalies, thereby achieving accurate detection of unstable approach events. The method was trained and tested using the real flight data recorder (FDR) data as an example, and it was found to be much better than existing multi-dimensional time series anomaly identification techniques that are appropriate for unstable approach detection. The F₁ score of the proposed method in this paper can reach 0.935, which is an average increase of 12.95% compared with other methods.

Through the use of intersatellite link data and ground regional networks, the BeiDou-3 system offers worldwide navigation, location, and timing services. Currently, the precision of the BeiDou-3 broadcast ephemeris orbit and clock offset products has reached a high level. However, due to insufficient satellite coverage from the ground regional network, the accuracy of the differential code bias (DCB) parameters broadcast by the satellites shows a significant gap compared to other navigation system parameters and post-processed precision products. In order to tackle these problems, this study examines how the accuracy of DCB parameter estimation in a regional monitoring network is affected by variables like the distribution of monitoring stations, multi-GNSS combinations, and ionospheric delay correction models. It proposes an optimal strategy for estimating DCB parameters within this network. Using the DCB products released by the Chinese Academy of Sciences (CAS) as a reference, the stability and consistency of the DCB products are analyzed, and their accuracy is evaluated through the signal in space range error (SISRE). Single-point positioning accuracy tests are conducted using BDS-3 B1I/B3I dual-frequency data. The findings show that, with a root mean square difference of roughly 0.4 ns, the satellite-end DCB stability in the regional monitoring network is roughly twice as poor as that of the satellite-end DCB determined using the CAS global monitoring network. The stability of satellite-end DCB calculated from dual-system observation data is superior, and the BDS-3 satellite DCB shows better consistency with the CAS DCB products. The receiver-end DCB in the regional monitoring network exhibits the same stability as the CAS DCB products, with dual-system observation data yielding better consistency for the receiver DCB. Compared to the SISRE results of BDS-3 calculated using TGD1 parameters from broadcast ephemeris, the regional monitoring network DCB products enhance the SISRE calculation accuracy for BDS-3 by 47.6%. The dual-frequency SPP positioning results improve by 14.1%, similar to the enhancement effect observed with the CAS DCB products.

Polar codes have become prevalent in the fifth-generation mobile communication technology (5G) owing to their capacity characteristics and straightforward compilation. The belief propagation (BP) decoding algorithm, which demonstrates parallel execution and a high throughput rate, is a commonly employed polar code decoding algorithm. This paper proposes a BP decoding algorithm based on task graph reconstruction (TGR) to reduce the algorithm's decoding complexity and delay. Using graph equivalence relations, the decoding algorithm is structurally optimized in two steps. Firstly, the redundancy elimination algorithm is used to remove the redundant calculations in the BP decoding algorithm and simplify the algorithm's operation structure. Subsequently, the branch transformation algorithm is used to optimize the operation order and reduce the critical path delay. The suggested BP algorithm has a better overall performance, particularly in application scenarios with sensitive delay and harsh channel conditions, than the three BP decoding algorithms that aim for structural optimization. It can reduce the critical path delay by at least 2.3%, reduce the computational complexity by at least 4.2%, and virtually eliminate the loss of error correction performance.

The oxygen concentrator is an important airborne part of the aircraft life support system, and its performance degradation data has the characteristics of multivariate and strong noise. In order to solve the problem of lack of univariate information and low prediction accuracy in the prediction of oxygen concentrator life, the correlation between multi-dimensional degradation variables was considered to select suitable variables. By introducing the influencing factor of oxygen concentration into the oxygen partial pressure variable, the long-term trend in the sequence was extracted with the Hodrick Prescott (HP) filter, and the fuzzy C-means (FCM) method was used to stage the oxygen partial pressure to establish a multi-stage dynamic regression model. Modeling the degradation of an oxygen concentrator. The findings indicate that the multivariate autoregressive integrated moving average (ARIMA) model with HP filtering improves prediction accuracy by 98.20%, 81.21%, and 77.87%, respectively, when compared to the univariate ARIMA model, single-stage multivariate ARIMA model, and multi-stage dynamic regression model.

Due to the complex texture, large redundant space, and widespread application of images, image based steganography algorithms are still the mainstream direction of steganography. Deep neural network-based picture steganography algorithms have become a research hotspot in the field of image steganography in recent years due to their increasingly good steganographic performance. This article proposes a high-capacity image steganography algorithm based on a generative adversarial network (GAN). The algorithm designed an information embedding and extraction network with a preprocessing module, a convolutional neural networks (CNN) module, and a U-Net as the main components. Through the constraint of the loss function, the embedding, extraction, and discrimination networks were jointly trained to achieve good steganographic visual effects. The experimental results show that our algorithm achieves an embedding capacity of 24 bpp. The overall superiority of the end-to-end steganography network designed in this paper is demonstrated by the fact that, under the assumption of high-capacity embedding, the encrypted images produced by the algorithm in this paper and the extracted secret images are higher than other comparable algorithms in both subjective visual quality and objective visual indicator peak signal-to-noise ratio (PSNR).

Accurate detection and exclusion of faulty measurements is the key for advanced receiver autonomous integrity monitoring (ARAIM) to ensure the safety of navigation service in the aviation field. With the deployment of new constellations such as Beidou and GLONASS, the number of fault modes to be monitored increases sharply, and it is difficult for traditional ARAIM to take into account the success rate and computational efficiency at the same time by using recursive search method. This paper presents a multi-constellation ARAIM fault exclusion method based on maximum a posteriori probability (MAP). On the one hand, this method realizes the risk control of misarrangement of healthy satellites by combining the prior probability information of faulty measurements. On the other hand, the posterior probability distribution model of faulty measurements under the hypothesis of potential failure mode is accurately constructed based on the maximum likelihood estimation to avoid the occurrence of incomplete exclusion events of faulty measurements. Additionally, this article incorporates the suggested MAP fault exclusion approach with the fault feature judgment of faulty measurements to create an ARAIM fault exclusion framework that enhances fault exclusion efficiency. Simulation and experimental results show that, compared with the traditional ARAIM fault exclusion method, the proposed method can improve ARAIM fault exclusion efficiency and ensure the accuracy of fault exclusion in both single-satellite and multi-satellite fault scenarios.

This paper examines the attitude and orbit tracking control problem of rigid spacecraft when there are uncertainties in both the irregular terms of the asteroid’s gravitational field and the spacecraft’s mass characteristic parameters. Based on the spacecraft’s prescribed performance error motion model described by Lie groups under irregular gravitational fields, a composite adaptive attitude and orbit tracking controller is proposed. A parameter update rule with higher convergence performance is devised based on the dynamic regression extension method and the immersion and invariance (I&I) theory to estimate the mass characteristic parameters with the goal of minimizing their uncertainty. Utilizing the estimated values of the mass characteristic parameters, an extended state observer is devised to estimate the overall system disturbance, which stems from the irregularity of the gravitational field and external disturbances. Building upon this parameter update law and disturbance estimation compensation, a composite adaptive prescribed performance terminal sliding mode controller is formulated. By using Lyapunov theory, it is demonstrated that the suggested controller guarantees that the errors in mass characteristic parameter estimate, disturbance estimation, and attitude and orbit tracking stay within a certain bound. Simulation results demonstrate that the incorporation of the dynamic regression extension method enhances the convergence performance of mass characteristic parameter estimation, and further, the addition of disturbance estimation compensation improves the accuracy of attitude and orbit tracking control.

To address the issue of inefficient refueling vehicle scheduling during airport flight support operations, this study establishes dual constraints for flight refueling time windows and permissible commencement time windows while prioritizing flight safety. It creates three integer programming models that optimize refueling vehicle allocation under these two limitations, standardizes stand designations and apron positions, and looks at key variables in three operational situations. According to the actual operation of the airport, a genetic optimization algorithm with elite strategy was proposed to solve the model according to the actual operation of the airport. In the first stage of independent scheduling, an evaluation mechanism that takes into account the two abilities of adaptability and population diversity was designed to ensure the diversity of the population. In the second and third stages of collaborative scheduling, a penalty factor introduction algorithm is designed, which can effectively avoid the emergence of populations that do not meet the target constraints. The simulation results show that compared with the traditional manual scheduling and genetic algorithm, the number of dispatched fuel trucks is reduced by 27.6% and 21.8% on average, which can provide effective decision support for airport fuel truck scheduling.

This paper propose a synchronous installation method for the integration of flexible bracket cables and photovoltaic modules, and design installation equipment, in response to the construction of a photovoltaic power station with a “Fishery-PV Integration” flexible bracket system. In order to investigate position and tension controllers for flexible supports, a dynamic model of the flexible support must be established in order to examine the features of the system. Additionally, a delay prediction method must be designed to account for the random delay in long-distance communication across water. The position and tension controls adjust for the anticipated delay. Through simulation and experimental verification, it has been found that the position error and tension error of photovoltaic module installation equipment based on a flexible support system are controlled within the range of 0.00645 m and 327.3 N, respectively, when considering the effects of random delay and processing errors. This proves the accuracy of the controller design and provides a new solution for the automation construction of photovoltaic power stations with a flexible support system.

A U-shaped network is suggested for anomaly behavior recognition based on Mamba in order to overcome difficulties in unsupervised anomaly behavior detection, such as the predictor’s propensity for abnormal generalization and the target objects' scale disparities. The network improves both global and local features to constrain undesired generalization ability in predictions. A state space model is introduced in the encoder to strengthen the extraction of global features. A multi-scale spatial channel fusion (M-SCF) strategy is designed to integrate feature information from different receptive fields, thereby reducing the interference of scale differences on local features. Skip connections are used in the decoder to enrich shallow feature information and enhance the ability to capture contextual information. The proposed method has been extensively validated on the UCSD Ped2, Avenue, and Shanghai Tech datasets, with respective recognition accuracies of 98.1%, 89.8%, and 78.5%. Mamba can successfully increase the accuracy of abnormal behavior identification, as evidenced by the findings, which demonstrate superior accuracy when compared to several sophisticated algorithms in recent years.