Search Results (1,117)

Recent advances in NeRF-based inpainting have enabled the completion of masked regions across multi-view images. However, two major challenges remain: generating accurate masks efficiently in the presence of complex multi-object interference and maintaining view consistency without floating artifacts in large-scale, unbounded 360° environments. To address these challenges, we propose a 3D inpainting framework with two principal components. First, a depth-aware mask-generation pipeline produces high-quality, view-consistent masks from sparse annotations. Second, spherical parameterization is combined with a joint objective comprising smooth inter-layer and original-scene constraints to suppress floating artifacts and content drift in unbounded scenes. We also introduce IM2360, a multi-object 360° dataset for evaluating 3D inpainting methods. Experiments show that our approach outperforms existing NeRF-based inpainting methods in PSNR, LPIPS, and FID. On a single NVIDIA GeForce RTX 4090, our method requires an average of 25.21 min of training per scene, compared with 98.14 min for SPIn-NeRF, 38.71 min for OR-NeRF, and 11.29 min for InFusion. These results indicate that the proposed method provides a favorable balance between reconstruction quality and computational cost for high-fidelity restoration of complex immersive scenes. Full article

As transmission systems evolve toward lightweight design, gear webs are becoming thinner and more sensitive to deformation caused by the coupled action of heat-treatment residual stress, finish-turning thermo-mechanical loading, and clamping constraints. Existing studies mainly treat the initial residual stress or the machining-induced residual stress separately and often simplify the clamping boundary as an ideal fixed constraint. To overcome these limitations, this study proposes an initial-field-driven prediction framework for aerospace thin-web gears. The post-heat-treatment residual stress/strain field is reconstructed using the eigenstrain reconstruction method using measured residual stress and deformation data and is then introduced into the ABAQUS finish-turning model as the actual initial state. A three-jaw-chuck boundary consistent with the experiment and a progressive element birth–death strategy driven by the measured cutting force and temperature are used to describe material removal. In addition, a laser displacement sensor on-machine measurement (LOMM) method is developed for initial pose correction, deformation monitoring, and clamping-force interval optimization. The predicted final residual stress (FRS) distribution and machining deformation agree with the experimental measurements, with errors generally below 10%. The optimized clamping-force interval of 1350–1650 N provides a balance between cutting stability and deformation suppression. This work clarifies the coupled roles of the initial heat-treatment state and clamping constraints in thin-web gear finish turning and provides a reproducible modeling route for FRS and deformation prediction. Full article

Against the backdrop of the rapid development of digital marketing in emerging economies, livestreaming e-commerce provides a new pathway for products from resource-constrained regions to overcome geographical limitations and expand market channels. Existing studies have largely focused on the sales performance, platform models, or general determinants of purchase intention in livestreaming e-commerce, while insufficient attention has been paid to the formation mechanism of consumers’ purchase intention within livestreaming interactions. Drawing on the S-O-R model and interaction ritual chain theory, this study constructs a theoretical model of how rural livestreaming e-commerce influences consumer purchase intention. Specifically, live streaming scenario atmosphere, product packaging, anchor interaction, and consumer engagement are identified as key stimulus factors. This study examines how these factors influence purchase intention through affective perception, cultural perception, and consumer trust, and further investigates the moderating role of emotional energy. The results show that: (1) the four livestreaming interaction factors significantly enhance consumers’ affective perception and cultural perception; (2) affective perception and cultural perception each form a chain mediation path with consumer trust, playing an important transmission role between livestreaming stimuli and purchase intention; and (3) emotional energy strengthens the effects of certain livestreaming stimuli on consumers’ psychological perceptions. This study reveals the pathway of “livestreaming stimuli–affective/cultural perception–consumer trust–purchase intention,” and provides strategic implications for livestreaming marketing, local brand communication, and rural economic development. Full article

Over the years, industrial demands have determined the main course of electric drives research and development. Modern drive trains are forced to provide extremely efficient operation under a variety of unfavorable circumstances. Moreover, the maintenance of the drive is often a critical factor, including both its reliability in the long-term perspective and deployment costs. In addition, the sophistication of up-to-date industrial machinery increases the number of stochastic disruptions that affect the final control quality. Thus, the Control Theory satisfies the need for a novel, robust strategy by proposing the Active Disturbance Rejection Control (ADRC) algorithm. It stands out with great dynamic performance and versatility. It has been widely tested in a variety of different industrial applications, including aviation, autonomous and unmanned vehicles, marine robots, automotive solutions, renewable energy, and power systems. Many of the above-mentioned applications use electric drive units. This paper elaborates on the review of the current state-of-the-art in the field of electric drive control with the ADRC strategy employed. Then, the ADRC designs regarding multi-mass drive trains are reviewed with emphasis on the speed control issue. This paper evaluates its variants and control approaches depending on the application purpose. Moreover, an exemplary dynamic properties analysis is performed to verify the default effectiveness of the algorithm. Then, the summary section is followed by an indication of possible future research directions. Full article

Many unsupervised domain adaptation methods require simultaneous access to both source-domain and target-domain data during the training stage. However, in practical electro-mechanical actuator (EMA) fault diagnosis, source-domain data may be unavailable, or it might not be able to be directly shared because of data ownership, confidentiality, or deployment constraints. Moreover, variations in command waveform, amplitude, load, and operating speed can cause distribution shifts between different working conditions, which degrades the diagnostic performance of models trained under a single condition. The domain adaptation model will also experience decreased performance in the source domain due to catastrophic forgetting. Hence, a dual-domain adaptation model is proposed, one which can reduce the risk of data leakage and relieve catastrophic forgetting in the source domain while improving domain adaptation in the target domain. The proposed framework transfers a source-trained model to the target domain without directly accessing source-domain samples. Sparse domain attention is introduced into the feature network to construct source-related and target-related feature channels, aiming to reduce source-domain performance degradation during target-domain adaptation. In addition, the classifier is fixed according to the source-domain hypothesis, and local structure clustering is used to improve pseudo-label consistency for unlabeled target-domain samples. The effectiveness of the proposed method is compared with several methods, utilizing low-frequency and high-frequency monitoring signals of the electro-mechanical actuator dataset, respectively. Full article

As uncertainty increases within the aviation industry, the importance of airport sustainability has come to the forefront, necessitating continuous demand generation through strategic aviation marketing. Consequently, there is an urgent need to analyze the practical efficacy of the incentive programs implemented primarily by global airports. In particular, amid intense competition among Northeast Asian airports to attract airlines, this study empirically analyzes the effects of Incheon International Airport’s (ICN) incentive programs on airline supply and passenger demand growth. To this end, utilizing empirical data from ICN spanning 2016 to 2024, this research quantitatively establishes the causal mechanism between incentive programs, supply expansion, and demand growth. Specifically, it comprehensively evaluates the impact of incentives provided for new airline entry and new route expansion, while comparing the effects across different airline business models. The empirical results confirm that the effects vary significantly depending on the type of incentive program and the airline category. Furthermore, the findings indicate that airport incentives are not merely sunk costs, but rather strategic investments that generate measurable aviation demand and airport revenue. However, the analysis also suggests a need to supplement and refine these incentive schemes to tailor them to specific airline types. Full article

High-energy-density fuels (HEDFs) have garnered considerable interest in aerospace fields, primarily due to their superior density and volumetric net heat of combustion (NHOC) compared with traditional petroleum-based fuels. Strained polycyclic hydrocarbons are regarded as one of the most crucial categories of HEDF. As an isomer (C₁₀H₁₆) of JP-10, the target compound is composed of two cyclopropyl rings and one cyclobutyl ring connected in a linear manner. Notably, intramolecular H-migration reactions of peroxyl radicals derived from strained polycyclic hydrocarbons (C₁₀H₁₅OO•) are of great significance for establishing the reaction mechanism of high-energy-density fuels over a broad temperature range. In this work, the intramolecular H-migration kinetics of C₁₀H₁₅OO• with distinct cis and trans configurations are investigated by quantum chemical calculations. Geometry optimization and frequency calculations are carried out for all species using the M06-2X/6-311++G(d,p) level of theory, while single-point energy calculations are performed at the CBS-QB3 level. Our calculated results demonstrate that different types of intramolecular H-migration reactions exhibit significant differences in barrier heights. Based on the ring structures where the reaction centers are located, these reactions can be classified into three categories: the lowest barriers correspond to H-migration reactions occurring between the central cyclopropyl ring and the terminal cyclobutyl ring; the highest barriers correspond to H-migration reactions confined entirely within the terminal cyclobutyl ring; and the barriers for H-migration reactions occurring between the terminal cyclopropyl ring and the central cyclopropyl ring lie between the above two. High-pressure-limit rate constants for 33 elementary reactions are determined in the temperature range of 500 to 2500 K based on the conventional transition-state theory (TST) and expressed in the modified Arrhenius form. Full article

The safe and accurate measurement of liquid hydrogen (LH₂) tank fill levels is a critical enabling technology for the adoption of hydrogen as a sustainable aviation fuel. Although LH₂ fill level measurement techniques have been applied in industrial, automotive, and space applications, no system has yet been validated at the scale, robustness, and precision required for modern aircraft Fuel Quantity Indication Systems (FQIS). Differentialpressure sensors are commonly employed in industrial cryogenic systems and hydrogen refueling stations; however, their accuracy is strongly influenced by dynamic effects such as filling transients and liquid sloshing, rendering them unsuitable for aviation-grade FQIS requirements which call for high accuracy and reliability. While simulations and analytical studies propose alternative LH₂ level sensing concepts, experimental validation and direct comparative assessments of different sensor architectures remain scarce. Furthermore, although several manufacturers offer LH₂ fill-level sensors, the stated measurement accuracies have not been independently verified, highlighting the need for systematic experimental investigation under representative operating conditions. A complete evaluation of an ${\mathrm{LH}}_2$ FQIS requires testing under anticipated flight conditions, including accelerations, varying attitudes, vibrations, dynamic sloshing, and long-term cycling. As a preliminary investigation, this work experimentally evaluates five liquid level sensing concepts based on measurements of dielectric constant, thermal capacity, and optical absorption properties using liquid nitrogen (${\mathrm{LN}}_2$) as a representative surrogate for ${\mathrm{LH}}_2$ under quasi-static conditions. The results demonstrate that optical absorption-based sensors in the near-infrared spectrum are unsuitable for LH₂ and LN₂ liquid level measurement. In contrast, capacitive probes and resistive thermal devices (RTDs) exhibit robust and repeatable performance under cryogenic conditions, demonstrating measurement resolutions of better than $5.1\mathrm{mm}$. These findings provide experimentally grounded guidance for the development of future LH₂-compatible FQIS architectures for aviation applications. Full article

In contrast to the development of conventional civil aircraft propulsion systems, novel propulsion technology or non-standard energy sources disclose a lack of flexibility in the current aviation certification framework. Additionally, aircraft certification in 2025 relies heavily on manual effort, causing major difficulties in maintaining the traceability of data across vast sets of regulation documents. One promising solution to overcome such challenges is offered by the field of artificial intelligence (AI), particularly large language models (LLMs). This paper introduces an AI-aided framework designed to streamline the certifiability of novel aircraft propulsion systems. To meet the AI trustworthiness demands set by the European Union Aviation Safety Agency (EASA), the framework proposes a concept for achieving data and model transparency in machine learning (ML) applications. To address the demand for data transparency, aviation regulatory context data will be unified and stored in a Unified Regulations Database (URD). This unified data is classified and enriched with related information for ML purposes. The URD enables the creation of modern, transparent AI features for civil aircraft certification. This AI-aided framework will enable certification measures for the development and allows for certifiability checks for novel aircraft technologies. Both the aviation industry and regulatory authorities may equally benefit from the existence of the URD as a starting point for certifying AI features. Full article

The aviation industry is adopting liquid hydrogen (LH₂) for sustainable flight, requiring robust safety systems. This work is an example of adaptation of a Micro-Electro-Mechanical Systems (MEMS)-based structural health monitoring (SHM) system for LH₂ tanks, developed in the H2ELIOS project. It uses a multisensor approach that combines MEMS sensors to monitor vibration and acceleration, shape memory alloy (SMA) strain sensors for measuring tank expansion, and hydrogen leakage sensors to prevent false alarms. This SHM technology detects cracks and delamination of material and coating, enabling predictive maintenance via digital twins and ensuring structural integrity. Full article

Municipal solid waste (MSW) management is a critical challenge to advancing recycling and circular economy approaches. This review provides a comprehensive overview of MSW management, encompassing sourcing, policy frameworks, characterization techniques, separation technologies, preprocessing strategies, and utilization pathways. First, generation patterns and sourcing mechanisms are discussed in both U.S. and global contexts, with emphasis on the influence of policy frameworks on waste reduction and diversion. Second, characterization techniques are evaluated, focusing on physical and chemical analysis for material recyclability. Third, sorting technologies are critically reviewed, covering conventional methods and emerging sensor-based approaches. Preprocessing techniques are then evaluated for their role in improving downstream conversion efficiency. Finally, valorization pathways such as waste-to-syngas, waste-to-biochar, and waste-to-sustainable aviation fuel (SAF) are assessed in terms of their role in climate mitigation and the circular economy. It is anticipated that this review will provide a foundational reference for researchers, policymakers, and industry stakeholders aiming to strengthen the recyclability infrastructure and maximize the efficiency of MSW management systems in the framework of the circular economy. Full article

Although many researchers use pre-trained models to better solve downstream tasks, further exploration of more effective pre-training methods remains necessary, especially for multi-modal pre-training where high-quality training data is more difficult to obtain. This work aims to improve the knowledge-learning performance in multi-modal pre-training. Some researchers focus on injecting entity knowledge into language pre-trained models based on masked entity model (MEM) training, which masks entities randomly and lets the model recover. These methods cannot guarantee good performance due to the lack of consideration of which entities are more valuable for learning. Moreover, in multi-modal training data, some entities may be unrelated to visual content. In this work, for the vision-language pre-trained model, we propose a Masked Entity Model pre-training method based on Active learning (ActiveMEM). It is designed to actively mask important and informative entities—those that are both informative and uncertain—for the model to recover, thereby encouraging it to extract more valuable knowledge from the data. The proposed method is evaluated using three pre-training datasets and four downstream datasets, and the experimental results demonstrate the effectiveness of our method. Full article

Against the backdrop of the dual-carbon strategy, China’s civil aviation industry, as a high-energy-consumption and high-carbon-emission sector, faces mounting pressure for low-carbon transformation. As the dominant airlines within China’s civil aviation system, Air China, China Eastern Airlines, and China Southern Airlines play a pivotal role in guiding the industry’s high-quality development. Employing the Global Malmquist–Luenberger (GML) index model, this study constructs a global production frontier incorporating undesirable outputs to systematically measure the dynamic evolution of total factor productivity (TFP) for the three major airlines in the period 2005–2023, and further applies a combined static-dynamic regression framework to identify the firm-level heterogeneous mechanisms through which explanatory factors operate. The results reveal significant heterogeneity in TFP trajectories: China Southern Airlines exhibits the most stable efficiency with the lowest volatility; China Eastern Airlines displays the greatest volatility but the strongest post-crisis rebound; and Air China occupies an intermediate position in both efficiency level and volatility. This differentiation stems from fundamental differences in market positioning, strategic orientation, and resource allocation patterns. Market competitiveness exerts a significantly positive effect on TFP for both Air China and China Eastern Airlines. Technological innovation investment generates short-run negative effects across all three airlines, albeit with divergent magnitudes. Human capital accumulation acts as a positive driver for Air China but produces a negative effect for China Southern Airlines, attributable to a structural mismatch between aggressive talent upgrading and organizational absorptive capacity. Shifting the unit of analysis to the firm level, this study identifies three heterogeneous strategic archetypes—market-led, scale-expansion, and regional-deepening—and constructs a differentiated “one firm, one policy” framework to provide targeted policy guidance for improving airline efficiency and facilitating low-carbon transition under carbon constraints. Full article

Autonomous driving systems (ADSs) are reliable only when heterogeneous sensors, estimation algorithms, and safety mechanisms are engineered as a single coherent safety-critical measurement system rather than as loosely coupled modules. Production stacks integrate cameras, LiDAR, automotive radar, and GNSS/IMU, yet deployment remains constrained by modality-specific failure modes, calibration and synchronization drift, and out-of-distribution (OOD) conditions that violate modeling assumptions. These limitations induce overconfidence and downstream decision errors whenever planning assumes certainty sharper than sensing can justify. This survey introduces a sensor-centric framework linking measurement physics, uncertainty propagation, fusion integrity, safety assurance, and risk-aware planning and control. We formalize what each modality physically measures; unify probabilistic, evidential, and conformal uncertainty representations; analyze filtering, factor-graph, BEV, transformer, and state-space fusion architectures with an emphasis on robustness and graceful degradation; and generalize aviation-style integrity concepts (RAIM/ARAIM) to multi-modal autonomy. The distinctive contribution is a single sensor-to-assurance throughline in which every uncertainty representation is tied to its measurement physics, every fusion architecture is evaluated against an explicit integrity-monitoring requirement generalized from RAIM/ARAIM, and every safety-standard clause is mapped to a concrete architectural mechanism. We map these mechanisms onto ISO 26262, ISO 21448 (SOTIF), ISO/PAS 8800, ANSI/UL 4600, and the UNECE framework, and connect perception uncertainty to decision-making through chance-constrained MPC and formal safety filters (RSS, CBF). Industry case studies and emerging V2X and generative-simulation approaches close the loop to deployable safety arguments. Full article

Four-phonon (4 ph) scattering is critically important for describing thermal transport properties in two-dimensional (2D) materials. Incorporating the 4 ph process is crucial for obtaining reliable lattice thermal conductivity (κ_l) and understanding phonon thermal transport. Among emerging 2D materials, monolayer BeN₄ has attracted increasing attention because of its unique structural properties. Here, the influence of 4 ph scattering on the thermal transport behavior of monolayer BeN₄ is comprehensively explored through first-principles calculations. The calculated results demonstrate that, after considering the 4 ph scattering, the κ_l of monolayer BeN₄ at 300 K are reduced by 37.7% and 50.6% along the zigzag and armchair directions, respectively. These findings indicate that monolayer BeN₄ exhibits anisotropy in thermal transport and that 4 ph scattering has a significant impact on thermal transport. The thermal transport is dominated by acoustic phonon branches. Furthermore, the larger κ_l at low temperatures originates from longer phonon lifetimes, larger phonon mean free paths, lower phonon scattering rates, and smaller weighted phase space. In addition, the different channels of 4 ph scattering are systematically analyzed, revealing that the redistribution channel provides the dominant contribution to 4 ph scattering. This investigation provides deeper insight into the thermal transport behavior of monolayer BeN₄ and facilitates its potential applications in nanoelectronic and thermal management devices. Full article