Search Results (2,151)

Deep learning models for power quality (PQ) disturbance classification remain critically vulnerable to adversarial perturbations, with classification performance degrading severely under white-box attacks. Existing defenses address individual models in isolation and provide no mechanism for operators to assess whether the system is under attack or which classifier remains trustworthy. This paper proposes a two-stage framework that combines adversarial training with large language model (LLM) reasoning to improve both robustness and interpretability. In the first stage, four architecturally diverse classifiers, including a proposed Multi-Scale Temporal Attention Network (MSTAN), are evaluated under four adversarial attacks (FGSM, PGD, C&W, and UAP), and their failure patterns are recorded as structured vulnerability fingerprints. The ensemble is then retrained via adversarial training on mixed clean and perturbed signals. In the second stage, an LLM analyzes the ensemble predictions alongside the fingerprint knowledge base to perform attack detection, fingerprint-guided meta-classification, and operator-facing threat report generation. On a 17-class, 255,000-signal synthetic benchmark, adversarial training recovers FGSM and PGD accuracy from below 25% to the 53–78% range, with MSTAN achieving the highest post-training robustness (78.26% under FGSM, 65.41% under PGD). The LLM reasoning layer provides an additional 3.5–6.2 percentage point improvement over majority voting by selecting the most reliable ensemble member based on the inferred attack condition, and detects adversarial attacks with 87.6% overall accuracy. To our knowledge, this is the first integration of LLM-based ensemble reasoning into the PQ adversarial robustness pipeline and the first application of the C&W optimization attack to power quality signals. Full article

Efficient management of soil–water resources is critical for global food security under intensifying climatic and demographic pressures. This review provides a comprehensive synthesis of artificial intelligence (AI) and distributed deep learning methodologies applied to soil–water interactions in precision agriculture. The physical and hydraulic foundations of soil–water systems—including water retention, unsaturated flow governed by the Richards equation, and soil degradation processes—are examined and situated within a unified framework of AI-based modeling and decision support. Classical machine learning (ML) algorithms (Random Forests, Support Vector Machines, gradient boosting) and deep learning architectures (convolutional neural networks, long short-term memory networks, transformers) are evaluated with respect to their capacity to predict soil moisture dynamics, estimate hydraulic properties, support smart irrigation scheduling, and generate digital soil maps at field-to-regional scales. Distributed training paradigms, federated learning for privacy-preserving multi-farm analytics, and edge AI deployment on low-power IoT hardware are assessed as enabling infrastructures for scalable agricultural intelligence. This review further addresses explainability, uncertainty quantification, and ethical dimensions inherent to AI-driven agricultural systems. Key challenges—including training data scarcity in data-poor regions, model interpretability, integration with physics-based hydrological models, and real-time deployment constraints—are critically discussed. Prospective research directions encompass physics-informed neural networks, foundation models for earth observation, autonomous digital twins of soil–water systems, and federated learning architectures aligned with data sovereignty frameworks. The synthesis underscores AI’s transformative potential for sustainable agricultural water management while delineating the technical and sociotechnical barriers that must be resolved to realize this potential at a global scale. Full article

Modern hands-free and wearable communication devices employ multiple microphones and loudspeakers, leading to the joint presence of acoustic echo, background noise, and desired speech signals. While acoustic echo cancellation (AEC) and beamforming are commonly combined to address this challenge, existing architectures face a trade-off between computational complexity, stability, and adaptability. In particular, adaptive beamforming approaches require repeated estimation and inversion of covariance matrices, incurring high computational cost and introducing potential sensitivity to time-varying conditions. Conversely, fixed beamformers reduce online complexity and improve stability, but their performance can degrade when the acoustic scene differs from the calibration condition. In this work, we investigate low-complexity AEC–beamforming architectures that combine fixed minimum-variance distortionless response (MVDR) beamforming with adaptive AEC. Since the ordering of these stages yields two inequivalent architectures, we evaluate two configurations: AEC followed by beamforming (AEC-BF) and beamforming followed by AEC (BF-AEC). To reduce dependence on a single head pose in wearable devices, we use an offline orientation-averaged calibration strategy in which the undesired-signal covariance matrix and, when required, the relative echo transfer functions (RETFs) are estimated from calibration measurements averaged across multiple head orientations. The proposed methods are evaluated using real-device recordings from a six-microphone wearable device. The results show a clear architectural asymmetry: the fixed BF-AEC configuration achieves the highest average echo return loss enhancement (ERLE) and perceptual evaluation of speech quality (PESQ), with substantially lower online complexity than the fully adaptive baseline, whereas the fixed AEC-BF configuration provides a higher signal-to-distortion ratio (SDR) in the evaluated experiment. Additional calibration experiments show that orientation-averaged RETF calibration provides partial generalization across the measured head orientations, but also that the RETFs are not fully orientation-invariant. Overall, the results indicate that fixed BF-AEC provides a favorable trade-off between echo suppression, stability, and online complexity under the evaluated real-recording conditions. Full article

[d=LE]Transitions of control between automated driving systems and human drivers remain safety-relevant and cognitively demanding moments in human–automation interaction. Recent studies show that transition performance depends not only on takeover timing or response speed but also on traffic complexity, driver readiness, automation limitations, trust calibration, and situational-awareness recovery. As in-vehicle interaction evolves toward conversational and agentic AI assistance, takeover support also becomes a problem of governing how natural-language AI systems communicate with the driver under uncertainty.Transitions of control between automated driving systems and human drivers remain safety-relevant and cognitively demanding moments in human-automation interaction. Recent studies suggest that transition performance should not be assessed only through takeover timing or response speed since control resumption quality also depends on traffic complexity, driver readiness, automation limitations, and situational awareness recovery. [d=LE]This paper proposes a digital-twin-mediated framework for human-aware takeover support in automated driving. In this framework, the companion AI is treated as an assumed LLM-based in-vehicle conversational or agentic assistant used as an advisory interaction component. The contribution is defined at the architectural level: human, vehicle, and context/road digital twins provide structured semantic state abstractions through a semantic state interface exposing confidence, freshness, provenance, and consistency metadata, while a trustworthy companion AI (TCAI) layer grounds, constrains, validates, and governs companion AI output proposals before HMI delivery.This paper motivates and defines a trustworthy companion AI (TCAI) layer for human-aware transition support in automated driving. The TCAI is conceived as a bounded, supervised, and explainable advisory agent that supports the driver without entering the safety-critical vehicle-control loop. It reasons over structured semantic state abstractions derived from a human digital twin, a vehicle digital twin, and a context/road digital twin, exposing driver readiness, automation capability, and contextual urgency in a form that supports traceable, uncertainty-aware, and degradation-aware assistance. [d=LE]Building on the research on driver-state monitoring, adaptive HMI, trust calibration, explainability, conversational assistance, and human assistance systems (HASs), the framework coordinates advisory interaction across vigilance support, contextual explanation, trust-calibrating communication, and directive handover guidance. The TCAI layer combines bounded reasoning, human-factor-derived guardrails, state-consistency management, dynamic explanation-depth control, trust-dynamics modeling, graded watchdog veto handling, mandatory access-control assumptions, and deterministic fallback. Safety-critical vehicle-control and minimum risk condition (MRC) functions remain assigned to the deterministic vehicle-control stack, while the authorized output path of the TCAI layer is validated HMI delivery.Building on the research on driver-state monitoring, adaptive HMI, trust calibration, explainability, and conversational assistance, we propose a conceptual architecture in which the TCAI coordinates multimodal assistance across different interaction conditions, including vigilance support, contextual explanation, trust-calibrating communication, and directive handover guidance. The companion does not actuate the vehicle; its outputs are constrained by runtime governance, policy enforcement, and deterministic fallback mechanisms. [d=LE]The paper concludes with a validation agenda and technical roadmap covering planned transitions, urgent handovers, degraded or adversarial conditions, temporal fusion of driver-state evidence, phase-sensitive HMI policies, trust-calibration trajectories, driver veto and partial-disabling mechanisms, and staged simulator-to-vehicle evaluation. Although motivated by SAE Level 3 automation, the framework may also inform fallback-related Level 4 scenarios in which human and automated agency must be managed under uncertainty.The paper concludes with a research roadmap for validating the proposed architecture under planned transitions, urgent handovers, and degraded or adversarial conditions. Although motivated by SAE Level 3 automation, the approach may also inform fallback-related Level 4 scenarios. Full article

Foggy traffic scenes pose significant challenges for object detection because reduced contrast, blurred object boundaries, and the loss of local details weaken discriminative feature representations. These degradations are particularly detrimental to lightweight detectors used in intelligent transportation and vehicle perception systems, where both accuracy and real-time efficiency are required. To address this problem, this paper proposes YOLO-GCM, a lightweight detector-side feature enhancement framework built upon YOLO11n. Instead of relying on an external image dehazing stage, YOLO-GCM improves the internal feature representation of the detector through three complementary modules: a gated additive feature block (GAFB) for adaptive channel-wise feature selection and noise suppression, a context-aware feature enhancement module (CAFEM) for strengthening high-level semantic context, and a multi-scale adaptive fusion (MSAF) module for enhancing cross-scale feature interaction. By integrating these modules into a unified one-stage detector, the proposed method improves detection robustness under low-visibility traffic conditions while maintaining a compact architecture. Experiments on the FoggyCar dataset show that YOLO-GCM achieved 89.81% mAP@0.5 and 67.99% mAP@0.5:0.95, outperforming standard YOLO baselines and dehazing-assisted detection pipelines under a consistent evaluation protocol. Additional evaluation on Foggy Cityscapes further verified the generalization capability of the proposed method under domain shift. The results demonstrate that detector-side feature enhancement provides an effective and efficient alternative to multi-stage dehazing-plus-detection pipelines for foggy traffic object detection. These findings can provide useful guidance for the development of robust and efficient perception modules in roadside monitoring, intelligent transportation systems, and vehicle-assisted driving applications under adverse weather conditions. Full article

Copper oxide (CuO) is a narrow-bandgap p-type semiconductor promising for visible-light photocatalysis, yet it suffers from rapid charge recombination and low carrier transfer efficiency. In this study, two distinct CuO photocatalysts were fabricated via different routes: two-dimensional CuO nanosheets derived from annealing a CuBDC metal–organic framework (MOF) precursor, and oriented one-dimensional CuO nanoflower arrays prepared by electrochemical deposition, followed by annealing. The crystal structure, morphology, optical absorption, and photoelectrochemical properties were systematically characterized by XRD, SEM, XPS, UV-Vis spectroscopy, transient photocurrent response, EIS, and PL spectroscopy. The CuO nanoflower thin film exhibits a broad visible-light absorption, a markedly higher photocurrent density (42.25 μA cm⁻²), and lower charge-transfer resistance compared to CuO nanosheets. When evaluated for visible-light photocatalytic degradation of methylene blue (MB), rhodamine B (RhB), and malachite green (MG), the CuO thin film completely degraded MB within 15 min, with an apparent rate constant of 20.15 h⁻¹—approximately three times that of CuO nanosheets. It also showed 1.2- and 1.28-fold higher activity for RhB and MG, respectively. The enhanced performance is attributed to the oriented nanoflower architecture that provides continuous charge transport pathways, suppresses carrier recombination, and extends light propagation via multiple reflections. This work demonstrates that microstructural engineering is an effective strategy to overcome the intrinsic limitations of CuO photocatalysts for wastewater treatment. Full article

As hen eggs are a primary source of high-quality dietary protein, egg freshness is fundamentally linked to biochemical alterations during storage, including moisture redistribution, protein degradation, and fluctuating chemical profiles. Accurate assessment of these internal changes is paramount for quality control; nonetheless, conventional analytical techniques remain predominantly destructive, rendering them impractical for high-throughput industrial monitoring. While existing literature has explored individual spectroscopic methods, the synergistic potential of multi-sensor integration and advanced artificial intelligence (AI) algorithms remains insufficiently synthesized. This review systematically evaluates recent breakthroughs in integrating AI with diverse spectroscopic modalities for non-destructive freshness quantification, including Visible-Near-Infrared (VIS-NIR), Raman, Fluorescence, and Hyperspectral Imaging (HSI). We elucidate the underlying mechanisms of spectral response to internal quality degradation and discuss the evolution of data-driven modeling from traditional chemometrics to sophisticated deep learning architectures. Furthermore, this work identifies critical bottlenecks in real-time industrial implementation and proposes future research trajectories toward intelligent multi-sensor fusion platforms. Full article

Cellulose is the most promising abundant renewable polymer material with the highest potential for the future low-carbon biorefineries. However, its utilization in industry is limited by the structural recalcitrance as a result of organization of crystalline domains, fibrillar architecture hierarchy and intramolecular and intermolecular hydrogen bonding which is responsible for access restriction for the catalysts and consequent cleavage of the glycosidic bonds. Therefore, efficient depolymerization of cellulose is of paramount importance as a step in biomass conversion into the low molecular products. In this review, the recent advances in cellulose depolymerization are discussed. The chemical, enzymatic, thermal, thermochemical, mechanochemical, oxidative and hybrid catalytic method is thoroughly discussed. Attention is paid to the mechanism of the depolymerization reaction steps as glycosidic bond activation as hydrolytic, radical mediated, and energy assisted pathways. Selectivity and conversion efficiency based on substrate morphology, solvent system and catalyst design are also discussed. Further, there is a comparison of key performance metrics which are relevant for the industrial process as product yield, carbon efficiency, energy demand, stability of the catalyst, solvent recyclability and impact to the environmental lifecycle. The pros and cons of the various methods are also represented. Processes based on mineral acids enable rapid conversion. However, they suffer from corrosion, waste handling issues and degradation by-products. On the other hand, enzymatic depolymerization processes offer relatively high selectivity but they are limited in terms of feedstock sensitivity and slow reaction kinetics. The downstream valorization mechanisms are also described with the result being that no single available technology is capable of satisfying all industrial requirements. Thus, future progress expects integrated circular processes where advanced catalysis, process intensification and digital optimization strategies take place. Full article

Fatigue health monitoring of engineering structures requires continuous degradation assessment, yet ground-truth health labels are unavailable during run-to-failure tests. Existing self-supervised approaches rely on monotonic degradation assumptions that are violated by the structured non-monotonic behaviour of acoustic emission signals during fatigue. A self-supervised framework called Cross-Modal Degradation Rivalry (CMDR) is proposed, which introduces the Modal Rivalry Index (MRI) as a directional measure of cross-modal predictability between heterogeneous sensor modalities. CMDR comprises a label-free representation-learning stage trained via the Cross-Modal Prediction Asymmetry (CMPA) pretext task, followed by a lightweight supervised stage that maps MRI features to scalar health indicators (HIs) using normalised lifecycle labels. The MRI is conceptually related, under the stated assumptions only loosely met in practice, to the Transfer Entropy difference between sensor latent channels. Experiments on a structural fatigue dataset with seven specimens under two loading conditions demonstrate that CMDR achieves competitive trendability and prognosability, as well as the lowest remaining useful life (RUL) error in three of four scenarios. RUL evaluations are additionally repeated under a fully online estimator that uses only training specimens. A strictly inductive ablation that re-pre-trains the self-supervised stage within each leave-one-specimen-out fold confirms a bounded transductive-vs-inductive gap, and CMDR remains the best against three further self-supervised baselines on the within-condition and mixed-condition scenarios. Ablation studies confirm the necessity of directional asymmetry, bottleneck architecture, and momentum-updated target encoders. Full article

Bone defects from trauma, tumour resection, infection, and degenerative disease remain a major clinical burden, and autografts face limitations of supply and donor-site morbidity. Three-dimensional (3D) printing offers a route to patient-specific, architecturally defined bone scaffolds, while biopolymers from natural sources provide biodegradability, biocompatibility, and extracellular matrix-mimicking cues consistent with sustainable, green biomaterials science. This review synthesises recent progress in 3D printing of biopolymer-based scaffolds for bone tissue engineering. We first examine the principal feedstocks—alginate, gelatin and gelatin methacryloyl, collagen, chitosan, silk fibroin, cellulose, and microbial polyesters—and their preparation, crosslinking chemistry, and printability. We then compare extrusion, light-based, and indirect printing technologies and the process–property relationships governing resolution, mechanical competence, and cell viability. Composite and functionalisation strategies, including biopolymer–bioceramic hybrids and controlled delivery of growth factors and antimicrobial agents, are analysed as routes to osteoinduction, vascularisation, and infection control. Finally, we evaluate translational performance in preclinical models and outline central challenges of vascularisation, mechanical–degradation matching, scalability, and regulatory standardisation. Biopolymer 3D printing is positioned as a ve rsatile, sustainable platform whose clinical maturation depends on integrated material, structural, and biological design. Full article

Accurate and reliable localization is a fundamental requirement for autonomous mobile robots operating in structured indoor environments. Adaptive Monte Carlo Localization (AMCL), widely used due to its probabilistic flexibility, suffers from performance degradation in challenging situations such as low-motion, sensor degradation, symmetry ambiguity, and abrupt position changes (kidnapped robot). This study proposes the Confidence-Gated Virtual Scan Motion AMCL (CG-VSM-AMCL) approach, which extends the standard AMCL structure with a selective and confidence-based posterior enhancement mechanism to overcome these limitations. The proposed method integrates beam partitioning, cluster-based dominance analysis, observability-aware gating, and recovery-driven adaptive particle injection components within a holistic architecture. The method was evaluated on a structured department map under seven representative scenarios: cold-start, low-motion, kidnapped robot recovery, odometry bias, scan dropout, world–model mismatch, and symmetry ambiguity. Experimental results demonstrate that the proposed approach systematically reduces localization error, false-lock rate, and convergence time compared to basic AMCL variants, and improves stability under challenging conditions. The significant improvements achieved, particularly in low-motion and symmetry-containing environments, reveal that selectively activated correction strategies can substantially increase localization robustness without altering the fundamental probabilistic structure of AMCL. Full article

Centrifugal compressors are critical assets in the oil and gas, petrochemical, and power generation industries, where unplanned downtime results in severe economic and safety consequences. Despite the application of digital twin technology for predictive maintenance, existing approaches struggle to combine accurate degradation modeling with formal assurance evidence that regulators and operators demand before trusting machine learning-augmented systems. This paper proposes a hybrid digital twin framework whose architecture is structured around a formal safety case template, addressing both the accuracy and the trustworthiness challenges simultaneously. The methodology couples a first-principles thermodynamic model with a neural-network residual learner, and the complete system is organized through a design-stage safety case constructed in Goal Structuring Notation. The design stage identifies the requirements for operational deployment. Validation through a simulation study on a one-year synthetic operational dataset shows that the hybrid model reduces root-mean-square prediction error by over 50% for both pressure ratio and polytropic efficiency compared to the physics-only baseline. The anomaly detection module, presented here as a proof of concept, achieves 92% recall in identifying injected faults, and a composite health index tracks the progression of fouling, erosion, and seal wear over the simulated service life. This study is purely theoretical, with no experimental measurements conducted. It demonstrates the structural viability and coherence of the proposed framework within a controlled environment, providing a solid theoretical and computational foundation for future physical validation efforts. These findings provide preliminary evidence that embedding a structured safety argument into the design of a hybrid digital twin is technically feasible and beneficial for building the confidence needed to deploy such systems in safety-critical industrial environments. Full article

Solar energy has emerged as a promising alternative to fossil fuels for mitigating climate change; however, efficient photovoltaic (PV) operation requires continuous monitoring and accurate energy forecasting. This study proposes an intelligent IoT-based photovoltaic monitoring and short-term energy prediction system integrating real-time sensing, solar tracking, and machine learning techniques. A small-scale experimental prototype based on a 10 W photovoltaic panel was implemented to collect real-time data, including voltage, current, temperature, humidity, ultraviolet radiation, and dust accumulation during a 30-day monitoring period under outdoor conditions. The acquired data were transmitted through an IoT architecture based on the Arduino Uno and ESP32, programmed using Arduino IDE, and integrated with the Blynk cloud platform for real-time monitoring and analysis. To evaluate predictive performance, Random Forest, XGBoost, and LSTM models were trained and compared for photovoltaic energy forecasting. Experimental results showed that XGBoost achieved the best predictive performance, obtaining the lowest error values (MAE = 0.00077, RMSE = 0.001103) and the highest coefficient of determination (R² = 0.918), outperforming the other evaluated models. In addition, the proposed system enabled effective remote monitoring and degradation analysis associated with environmental conditions. The results demonstrate the potential of integrating IoT and machine learning for accurate short-term photovoltaic energy forecasting in small-scale experimental environments. Nevertheless, further long-term and large-scale validation is required to evaluate system robustness under operating conditions. Full article

Network latency remains a fundamental bottleneck in vehicle teleoperation, inducing instability and performance degradation in conventional control methods, while predictive techniques like the Smith Predictor offer a theoretical solution, their efficacy is often compromised by unmodelled dynamics and real-world disturbances. This paper presents the first experimental validation of the Successive Reference-Pose Tracking (SRPT) architecture. By streaming future reference poses rather than direct steering commands, SRPT leverages an onboard Nonlinear Model Predictive Controller to compute optimal vehicle actions while inherently accounting for dynamic constraints and network delays. Real-world human-in-the-loop experiments were conducted with four drivers on a test track featuring cornering, double lane-change, and slalom manoeuvres. Quantitative comparisons at 10 km/h across four modes—manual driving, direct teleoperation, a Smith Predictor, and SRPT—demonstrate that SRPT significantly outperforms other teleoperation methods, reducing cross-track error by up to 66% and yielding smoother, more stable control inputs. Furthermore, SRPT uniquely maintained stability during a proof-of-concept trial at 13 km/h, where it proactively moderated vehicle speed to respect actuator limits—a critical safety behavior absent in other modes. This work provides the first tangible evidence that SRPT is a robust and superior framework for delay-resilient vehicle teleoperation in real-world conditions. Full article

In cold, humid environments, even a small amount of ice accumulation on the blade surface can degrade aerodynamic performance, increase drag, induce premature stall and vibration, and raise the risks of shutdown, fatigue, and ice throw. Existing blade anti-icing and de-icing strategies (such as passive coatings, electrothermal heating, hot-air systems, and hybrid designs) struggle to simultaneously meet the requirements of lightweight construction, low-voltage rapid heating, conformability to curved surfaces, erosion resistance, long-term durability, and scalable manufacturing. MXenes, particularly Ti₃C₂T_x, have attracted attention due to their high electrical conductivity, broadband optical absorption, solution processability, tunable interfacial chemistry, and good compatibility with polymer matrices. However, their oxidation issue and blade-scale deployment challenges (coating chemistry, scalable fabrication, real-world testing) remain obstacles. Based on this, this review discusses Ti₃C₂T_x-based anti-icing, de-icing, and defogging strategies for wind turbine blades, with emphasis on material properties, functional mechanisms, coating architectures, fabrication routes, durability, and scalability, and highlights their potential for lightweight and energy-efficient all-weather blade protection. Finally, future research directions for Ti₃C₂T_x-based blade anti-icing and de-icing are prospected. This review not only aims to identify key knowledge gaps in current research but also strives to provide a theoretical reference for the application of Ti₃C₂T_x in the complex service environment of real wind turbine blades, thereby moving beyond idealized laboratory conditions. Full article