Saved Queries

This comprehensive review examines hierarchical control principles and frameworks for grid-connected microgrids operating in environments prone to load shedding and under demand response. The particular emphasis is on South Africa’s current electricity grid issues, experiencing regular planned and unplanned outages, due to numerous factors including ageing and underspecified infrastructure, and the decommissioning of traditional power plants. The study employs a systematic literature review methodology following PRISMA guidelines, analysing 127 peer-reviewed publications from 2018–2025. The investigation reveals that conventional microgrid controls require significant adaptation to address the unique challenges brought about by scheduled power outages, including the need for predictive–proactive strategies that leverage known load-shedding schedules. The paper identifies three critical control layers of primary, secondary, and tertiary and their modifications for resilient operation in environments with frequent, planned grid disconnections alongside renewables integration, regular supply–demand balancing and dispatch requirements. Hybrid optimisation approaches combining model predictive control with artificial intelligence show good promise for managing the complex coordination of solar–storage–diesel systems in these contexts. The review highlights significant research gaps in standardised evaluation metrics for microgrid resilience in load-shedding contexts and proposes a novel framework integrating predictive grid availability data with hierarchical control structures. South African case studies demonstrate techno-economic advantages of adapted control strategies, with potential for 23–37% reduction in diesel consumption and 15–28% improvement in battery lifespan through optimal scheduling. The findings provide valuable insights for researchers, utilities, and policymakers working on energy resilience solutions in regions with unreliable grid infrastructure. Full article

(This article belongs to the Section A1: Smart Grids and Microgrids)

►▼ Show Figures

Figure 1

27 pages, 2126 KB

Open AccessArticle

Research on Fault Location Methods for Multi-Terminal Multi-Section Overhead Line–Cable Hybrid Transmission Lines

by Peilin Xu and Ruyan Zhou

Processes 2026, 14(3), 438; https://doi.org/10.3390/pr14030438 - 26 Jan 2026

Abstract

To address the fault location problem in multi-terminal hybrid overhead–cable transmission lines with multiple sections, this paper proposes a novel method combining Modified Ensemble Empirical Mode Decomposition (MEEMD) and the Teager Energy Operator (TEO). First, the MEEMD algorithm—which mitigates mode mixing—is integrated with the TEO, which captures instantaneous energy variations, to achieve accurate detection of traveling wavefronts. Considering the topological complexity of multi-terminal hybrid transmission lines, a fault branch separation and iterative judgment method is proposed. Based on the arrival time characteristics of traveling waves, two topology decoupling strategies are designed to enable branch identification through network reconstruction and iterative computation. After determining the faulted branch, the fault section is precisely localized by comparing the time difference between the arrival of traveling waves at branch terminals and T-nodes with the propagation time differences at each connection point. Finally, the dual-ended traveling wave method is applied to calculate the fault distance. The proposed method is validated through co-simulation using PSCAD 4.6.2 and MATLAB R2023b. Comparative analysis of ranging accuracy demonstrates that this approach ensures reliable fault location under varying fault positions and transition resistances. Full article

(This article belongs to the Section Energy Systems)

28 pages, 1133 KB

Open AccessReview

α,β-Unsaturated (Bis)Enones as Valuable Precursors in Innovative Methodologies for the Preparation of Cyclic Molecules by Intramolecular Single-Electron Transfer

by Tommaso Benettin, Francesca Franco, Fabrizio Medici, Sergio Rossi and Alessandra Puglisi

Molecules 2026, 31(3), 430; https://doi.org/10.3390/molecules31030430 - 26 Jan 2026

Abstract

The synthesis of monocyclic and bicyclic compounds plays a fundamental role in organic chemistry, and the need for novel synthetic methodologies is still under investigation. In particular, α,β-unsaturated (bis)enones have emerged as valuable precursors for the formation of cyclic (both mono and bicyclic) structures through single-electron transfer (SET) processes. Single-electron transfer (SET) is a redox process where one electron moves from a donor species to an acceptor, generating radical ions or neutral radicals that drive unique reaction pathways. Thanks to the advent of radical chemistry, it was possible to discover an entirely new reactivity of α,β-unsaturated (bis)enones, which, after a SET event, undergo the formation of cyclic molecules, both in intra and inter-molecular reactions, under several possible pathways, including formal [2+2] cycloaddition reaction (22CA) and 5-exo-trig cyclization, for ring closure. Today, the generation of radical species can be broadly classified into three main approaches: photochemical and photocatalytic, metal-driven and electrochemical processes. In this review, we summarize the progress achieved to date in the synthesis of cyclic molecules from α,β-unsaturated (bis)enones via single-electron transfer events under these three main classes of processes. Whenever possible, the reaction pathway and fate of the radical species generated through SET is discussed. Full article

(This article belongs to the Section Organic Chemistry)

23 pages, 3475 KB

Open AccessArticle

YOLO-GSD-seg: YOLO for Guide Rail Surface Defect Segmentation and Detection

by Shijun Lai, Zuoxi Zhao, Yalong Mi, Kai Yuan and Qian Wang

Appl. Sci. 2026, 16(3), 1261; https://doi.org/10.3390/app16031261 - 26 Jan 2026

Abstract

To address the challenges of accurately extracting features from elongated scratches, irregular defects, and small-scale surface flaws on high-precision linear guide rails, this paper proposes a novel instance segmentation algorithm tailored for guide rail surface defect detection. The algorithm integrates the YOLOv8 instance segmentation framework with deformable convolutional networks and multi-scale feature fusion to enhance defect feature extraction and segmentation performance. A dedicated guide rail surface Defect (GSD) segmentation dataset is constructed to support model training and evaluation. In the backbone, the DCNv3 module is incorporated to strengthen the extraction of elongated and irregular defect features while simultaneously reducing model parameters. In the feature fusion network, a multi-scale feature fusion module and a triple-feature encoding module are introduced to jointly capture global contextual information and preserve fine-grained local defect details. Furthermore, a Channel and Position Attention Module (CPAM) is employed to integrate global and local features, improving the model’s sensitivity to channel and positional cues of small-target defects and thereby enhancing segmentation accuracy. Experimental results show that, compared with the original YOLOv8n-Seg, the proposed method achieves improvements of 3.9% and 3.8% in Box and Mask mAP₅₀, while maintaining a real-time inference speed of 148 FPS. Additional evaluations on the public MSD dataset further demonstrate the model’s strong versatility and robustness. Full article

(This article belongs to the Special Issue Deep Learning-Based Computer Vision Technology and Its Applications)

►▼ Show Figures

Figure 1

23 pages, 1005 KB

Open AccessReview

Advances in Novel Biologics Targeting BAFF/APRIL in the Treatment of IgA Nephropathy

by Yiduo Xu, Yingqiu Mo and Youhua Xu

Cells 2026, 15(3), 240; https://doi.org/10.3390/cells15030240 - 26 Jan 2026

Abstract

IgA nephropathy (IgAN) is the most common primary chronic glomerular disease worldwide. Its clinical features include proteinuria and complement pathway activation, which are the strongest predictors of progression to renal failure. This disease can occur at any age. Approximately 30–40% of IgAN patients progress to end-stage renal disease (ESRD) within 20–25 years after diagnosis, making it one of the major causes of ESRD. As understanding of the autoimmune development of IgA nephropathy (IgAN) grows, research shows that BAFF and APRIL promote B-cell activation by binding to the receptors TACI, BCMA, and BAFF-R. This results in the overproduction of galactose-deficient IgA1 (Gd-IgA1), which helps drive the progression of IgA nephropathy. B-cell and plasma cell-targeted therapies, such as biologics against BAFF/APRIL, can precisely and effectively improve patient symptoms. Corresponding agents have now been successfully developed and are administered via subcutaneous or intravenous injection. Clinical trials have demonstrated the significant effectiveness of this approach, especially in reducing proteinuria, stabilizing eGFR, and lowering Gd-IgA1 levels. Although current trial data for BAFF/APRIL-targeted biologics in IgA nephropathy are promising, these new treatments need ongoing clinical monitoring for long-term infection risks and potential drug resistance. This article focuses on the application of BAFF/APRIL biologics in the treatment of IgA nephropathy, addressing gaps in existing literature. While prior studies have emphasized the mechanisms of action of these drugs in IgA nephropathy, they have lacked a comprehensive summary of the current status of specific drug research and clinical progress. Full article

►▼ Show Figures

Figure 1

17 pages, 787 KB

Open AccessArticle

Urinary Chemokines in the Diagnosis and Monitoring of Immune Checkpoint Inhibitor-Associated Nephritis

by Francisco Gomez-Preciado, Laura Martinez-Valenzuela, Paula Anton-Pampols, Xavier Fulladosa, María Jove, Ernest Nadal, Josep María Cruzado, Joan Torras and Juliana Draibe

Int. J. Mol. Sci. 2026, 27(3), 1240; https://doi.org/10.3390/ijms27031240 - 26 Jan 2026

Abstract

Immune checkpoint inhibitors are essential treatments for many oncologic diseases, but with well-known immune-related adverse events, such as acute interstitial nephritis (ICI-AIN). We investigated novel potential biomarkers that could assist in the diagnosis and follow-up of this condition and that are related to the active pathogenic pathways involved. We measured urinary soluble PD-1, PD-L1 and PD-L2, as well as chemokines CXCL5, CXCL9, CXCL10, CXCL11, CCL2, CCL3, CCL5 and cytokines IL-6 and IL-12p70 performing a Luminex assay in urine from patients with ICI-AIN (n = 35) and compared them with patients with AIN from other causes (non-ICI AIN) (n = 29) and ATN (n = 26). We found that CXCL5, CXCL9, CXCL10, CXCL11, CCL5 and IL-6 were higher in patients with ICI-AIN than in those with ATN, and all of them but CXCL9 and IL-6 were also higher in patients with ICI-AIN compared with non-ICI AIN. We also determined these molecules at follow-up for ICI-AIN patients (40 samples from 22 patients) and found that concentrations of CXCL9, CXCL10, CXCL11 and CCL2 decreased after treatment. The decrease of CXCL9 and CXCL10 correlated with greater kidney function recovery at one-year follow-up. These molecules could serve as noninvasive biomarkers and may aid fine patient monitoring. Full article

(This article belongs to the Special Issue Advances in Primary Nephropathies: Pathogenesis, Inflammatory Response and Advanced Treatments)

25 pages, 2201 KB

Open AccessArticle

Design and Research of a Dual-Target Drug Molecular Generation Model Based on Reinforcement Learning

by Peilin Li, Ziyan Yan, Yuchen Zhou, Hongyun Li, Wei Gao and Dazhou Li

Inventions 2026, 11(1), 12; https://doi.org/10.3390/inventions11010012 - 26 Jan 2026

Abstract

Dual-target drug design addresses complex diseases and drug resistance, yet existing computational approaches struggle with simultaneous multi-protein optimization. This study presents SFG-Drug, a novel dual-target molecular generation model combining Monte Carlo tree search with gated recurrent unit neural networks for simultaneous MEK1 and mTOR targeting. The methodology employed DigFrag digital fragmentation on ZINC-250k dataset, integrated low-frequency masking techniques for enhanced diversity, and utilized molecular docking scores as reward functions. Comprehensive evaluation on MOSES benchmark demonstrated superior performance compared to state-of-the-art methods, achieving perfect validity (1.000), uniqueness (1.000), and novelty (1.000) scores with highest internal diversity indices (0.878 for IntDiv1, 0.860 for IntDiv2). Over 90% of generated molecules exhibited favorable binding affinity with both targets, showing optimal drug-like properties including QED values in [0.2, 0.7] range and high synthetic accessibility scores. Generated compounds demonstrated structural novelty with Tanimoto coefficients below 0.25 compared to known inhibitors while maintaining dual-target binding capability. The SFG-Drug model successfully bridges the gap between computational prediction and practical drug discovery, offering significant potential for developing new dual-target therapeutic agents and advancing AI-driven pharmaceutical research methodologies. Full article

(This article belongs to the Special Issue Advances and Innovations in Deep Learning: Unveiling Multidisciplinary Applications and Challenges)

►▼ Show Figures

Figure 1

18 pages, 3671 KB

Open AccessArticle

Physiological Changes and Transposition of Insertion Sequences in the dps-Double-Knockout Mutant of Deinococcus geothermalis

by Yujin Park, Hyun Hee Lee, Eunjung Shin, Soyoung Jeong and Sung-Jae Lee

Int. J. Mol. Sci. 2026, 27(3), 1238; https://doi.org/10.3390/ijms27031238 - 26 Jan 2026

Abstract

DNA-protecting proteins (Dps) are crucial for safeguarding chromosomal DNA in starved cells during the stationary phase under stressful conditions. In previous research, the two Dps proteins in Deinococcus geothermalis, Dgeo_0257 (Dps3) and Dgeo_0281 (Dps1), were found to complement each other in protecting DNA from oxidative damage. This study investigates the physiological changes and transposition of insertion sequences (ISs) in a double-knockout (DK) mutant lacking both dps genes. Comparisons between the wild-type and mutant strains revealed significant phenotypic differences in viability under oxidative stress conditions induced by hydrogen peroxide and ferrous ions, particularly during the stationary phase. Notably, oxidative stress triggered the transposition of the IS families IS701 and IS5, with IS66 being transposed exclusively in the DK mutant into a gene encoding phytoene desaturase. Transcriptomic analysis using RNA-seq revealed substantial fold changes in gene expression across the genome. For example, the dgeo_1459–1460 gene cluster, which encodes a DUF421 domain-containing protein and a hypothetical protein, was highly upregulated under both oxidative and non-oxidative conditions. Interestingly, catalase, encoded by a single gene in D. geothermalis, was upregulated in the DK mutant during the stationary phase, with expression levels exceeding those observed in the single dps gene-deficient mutants. Conversely, a prominent downregulation of the Fur family regulator was detected. These findings highlight the growth phase-dependent physiological adaptation of the dps-DK mutant and reveal a novel IS transposition event of the ISBst12 group involving the IS66 family. Therefore, this study provides new observations into the influence of DNA-protective protein deficiency on oxidative stress responses and IS transposition in D. geothermalis, as well as the regulatory mechanisms of the catalase induction pathway, raising the need for further investigation into the role of OxyR. Full article

(This article belongs to the Section Molecular Microbiology)

►▼ Show Figures

Figure 1

13 pages, 1026 KB

Open AccessArticle

A Method to Determine the Habit Plane of a Dislocation Loop

by Yufeng Du, Lijuan Cui, Xunxiang Hu and Farong Wan

Materials 2026, 19(3), 497; https://doi.org/10.3390/ma19030497 - 26 Jan 2026

Abstract

The nature of dislocation loops significantly influences their evolutionary behavior and, consequently, affects the material properties, particularly under irradiation conditions. Determining the habit plane of a dislocation loop is the key point to examining its nature using the inside–outside method. In the present study, we introduce a novel technique for determining the habit planes of dislocation loops in the transmission electron microscope (TEM). The traditional inside–outside technique requires an edge-on perspective of the dislocation loop for analysis of the habit plane. In contrast, our innovative method for the precise determination of the habit plane delves into the geometric correlations between the dislocation loop and its projections under different crystal zone axes in TEM without being bound by the restrictive requirement of an edge-on view. It also simplifies the procedure of the inside–outside method. Furthermore, we have discussed the advantages and limitations of various methodologies employed to examine the nature of dislocation loops, as well as the techniques for determining their habit planes. Full article

22 pages, 31480 KB

Open AccessArticle

Bayesian Inference of Primordial Magnetic Field Parameters from CMB with Spherical Graph Neural Networks

by Juan Alejandro PintoCastro, Héctor J. Hortúa, Jorge Enrique García-Farieta and Roger Anderson Hurtado

Universe 2026, 12(2), 34; https://doi.org/10.3390/universe12020034 - 26 Jan 2026

Abstract

Deep learning has emerged as a transformative methodology in modern cosmology, providing powerful tools to extract meaningful physical information from complex astronomical data. This paper implements a novel Bayesian graph deep learning framework for estimating key cosmological parameters in a primordial magnetic field (PMF) cosmology from simulated Cosmic Microwave Background (CMB) maps. Our methodology utilizes DeepSphere, a spherical convolutional neural network architecture specifically designed to respect the spherical geometry of CMB data through HEALPix pixelization. To advance beyond deterministic point estimates and enable robust uncertainty quantification, we integrate Bayesian Neural Networks (BNNs) into the framework, capturing aleatoric and epistemic uncertainties that reflect the model confidence in its predictions. The proposed approach demonstrates exceptional performance, achieving

R^{2}

scores exceeding 89% for the magnetic parameter estimation. We further obtain well-calibrated uncertainty estimates through post hoc training techniques including Variance Scaling and GPNormal. This integrated DeepSphere-BNNs framework delivers accurate parameter estimation from CMB maps with PMF contributions while providing reliable uncertainty quantification, enabling robust cosmological inference in the era of precision cosmology. Full article

(This article belongs to the Section Astroinformatics and Astrostatistics)

30 pages, 2101 KB

Open AccessArticle

Empowering IoV Security: A Novel Secure Cryptographic Algorithm (OpCKEE) for Network Protection in Connected Vehicles

by Sahar Ebadinezhad and Pierre Fabrice Nlend Bayemi

Sensors 2026, 26(3), 825; https://doi.org/10.3390/s26030825 - 26 Jan 2026

Abstract

According to Fortune Business Insights, the market share of the Internet of Vehicless is expected to grow from USD 95.62 billion in 2021 to USD 369.61 billion in 2028, at a compound annual growth rate of 21.4%. However, the Internet of Vehicles system still faces several challenges, including regulation, scalability, data management, connectivity, interoperability, privacy, and security. To improve communication security within the Internet of Vehicle system, we have implemented a secure cryptographic algorithm called Optimized Certificateless Key-Encapsulated Encryption, resulting from a fusion of the key-insulated cryptosystem and the cryptographic key-encapsulated mechanism. The formal security analysis of our algorithm using the AVISPA version 1.1 software shows us that our protocol is safe. Informal analysis shows that our algorithm ensures authenticity, confidentiality, integrity, and non-repudiation and resists several other attacks. Our algorithm’s computational and communicational costs are slightly better than those at which it inherits the functionalities. Full article

(This article belongs to the Special Issue 5G/6G-Enabled IoT: Connectivity, Scalability and Low-Latency Applications)

►▼ Show Figures

Graphical abstract

17 pages, 2143 KB

Open AccessArticle

Combined Analytical and Clinical Performance Evaluation of a Novel Dengue NS1 Rapid Test in a Real-World Endemic Setting

by Jidapa Szekely, Hafik Duereh, Jenureeyah Mongkolprasert, Chadarat Senorit, Wilai Pattoom, Rawadee Suebsaiorn, Sirinda Woraphan and Piyawut Swangphon

Diagnostics 2026, 16(3), 395; https://doi.org/10.3390/diagnostics16030395 - 26 Jan 2026

Abstract

Objectives: This study evaluated the analytical and clinical performance of a novel NS1 rapid diagnostic test in a dengue-endemic setting in Thailand. Methods: The K-Dengue NS1 Ag test (K.Bio Sciences, Pathumthani, Thailand) was developed. Analytical performance included determination of LOD, reproducibility, and evaluation against potentially cross-reactive pathogens and interfering substances. Unlike conventional assays employing 40 nm colloidal gold, this test incorporates 80 nm gold nanospheres to enhance detection sensitivity. The LOD was determined by serial dilution of recombinant NS1 proteins representing all four dengue virus serotypes. Clinical performance was assessed using 185 archived plasma samples collected between January 2024 and February 2025 from two tertiary care hospitals in Thailand, with a commercial NS1 ELISA serving as the reference standard. Results: The K-Dengue NS1 test demonstrated serotype-specific limits of detection (LODs) for recombinant NS1 antigen, 2.9 ng/mL (DENV-1), 0.5 ng/mL (DENV-2), 25.2 ng/mL 27 (DENV-3), and 4.5 ng/mL (DENV-4). Cross-reactivity testing revealed no false positives against closely related arboviruses or common co-infections, and no interference was observed from frequently encountered pathogens or biochemical substances. In clinical evaluation, the assay achieved a sensitivity of 98.08% (51/52), a specificity of 100% (133/133), and an overall accuracy of 99.37%. Importantly, sensitivity was significantly higher in primary infections (100.00%) than in secondary infections (93.3%, p = 0.288). Conclusions: In this clinically oriented evaluation, the K-Dengue NS1 rapid test showed high specificity and good sensitivity, particularly in primary dengue infections. While the assay may be useful as part of early diagnostic workflows in comparable healthcare settings, reduced sensitivity in secondary infections indicates that negative NS1 results should be interpreted with caution and, where appropriate, supplemented with additional diagnostic methods. Full article

(This article belongs to the Special Issue Recent Advances in Epidemiological Diagnostics: Detecting and Controlling Infectious Diseases)

45 pages, 1611 KB

Open AccessArticle

Hidden Ethnomedicinal Diversity in a Fine-Scale Study from Konak, Eastern Anatolia

by Turgay Kolaç, Narin Sadikoğlu and Mehmet Sina İçen

Plants 2026, 15(3), 383; https://doi.org/10.3390/plants15030383 - 26 Jan 2026

Abstract

This study documents the ethnomedicinal knowledge of Konak (Malatya, Eastern Anatolia, Türkiye), a region with rich plant diversity but no prior comprehensive research. The aim of the study is to systematically document and analyze the ethnomedicinal practices of Konak village, focusing on plant taxa (species, subspecies and varieties) used, preparation methods, and therapeutic applications. Data were collected through semi-structured interviews with 68 local informants. Quantitative analysis was performed using Informant Consensus Factor (F_IC) and Use Value (UV) indices. Plant specimens were collected, identified, and deposited in the herbarium. The study documented 86 plant taxa from 35 families used in 230 therapeutic applications. Lamiaceae, Asteraceae, and Rosaceae were the most represented families. High F_IC values were recorded for colds (F_IC = 0.95), stomach pain (F_IC = 0.92), and inflammation (F_IC = 0.90), indicating strong community consensus. The most frequently cited species were Origanum vulgare subsp. gracile, Mentha spp., and Rosa canina. There are novel or locally specific uses, with 13 taxa having no previously recorded ethnomedicinal applications in the reviewed literature. The findings reveal Konak as a significant repository of ethnomedicinal knowledge. High-F_IC taxa represent prime candidates for phytochemical and pharmacological research to validate traditional uses and support evidence-based phytotherapy. This study enriches regional ethnopharmacological data and highlights candidate taxa for pharmacological validation. Full article

(This article belongs to the Section Plant Systematics, Taxonomy, Nomenclature and Classification)

20 pages, 1908 KB

Open AccessArticle

Research on Real-Time Rainfall Intensity Monitoring Methods Based on Deep Learning and Audio Signals in the Semi-Arid Region of Northwest China

by Yishu Wang, Hongtao Jiang, Guangtong Liu, Qiangqiang Chen and Mengping Ni

Atmosphere 2026, 17(2), 131; https://doi.org/10.3390/atmos17020131 - 26 Jan 2026

Abstract

With the increasing frequency extreme weather events associated with climate change, real-time monitoring of rainfall intensity is critical for water resource management, disaster warning, and other applications. Traditional methods, such as ground-based rain gauges, radar, and satellites, face challenges like high costs, low resolution, and monitoring gaps. This study proposes a novel real-time rainfall intensity monitoring method based on deep learning and audio signal processing, using acoustic features from rainfall to predict intensity. Conducted in the semi-arid region of Northwest China, the study employed a custom-designed sound collection device to capture acoustic signals from raindrop-surface interactions. The method, combining multi-feature extraction and regression modeling, accurately predicted rainfall intensity. Experimental results revealed a strong linear relationship between sound pressure and rainfall intensity (r = 0.916, R² = 0.838), with clear nonlinear enhancement of acoustic energy during heavy rainfall. Compared to traditional methods like CML and radio link techniques, the acoustic approach offers advantages in cost, high-density deployment, and adaptability to complex terrain. Despite some limitations, including regional and seasonal biases, the study lays the foundation for future improvements, such as expanding sample coverage, optimizing sensor design, and incorporating multi-source data. This method holds significant potential for applications in urban drainage, agricultural irrigation, and disaster early warning. Full article

(This article belongs to the Special Issue The Response of Plateau Vegetation to Climatic and Anthropogenic Drivers)

►▼ Show Figures

Figure 1

16 pages, 2052 KB

Open AccessArticle

Modeling Road User Interactions with Dynamic Graph Attention Networks for Traffic Crash Prediction

by Shihan Ma and Jidong J. Yang

Appl. Sci. 2026, 16(3), 1260; https://doi.org/10.3390/app16031260 - 26 Jan 2026

Abstract

This paper presents a novel deep learning framework for traffic crash prediction that leverages graph-based representations to model complex interactions among road users. At its core is a dynamic Graph Attention Network (GAT), which abstracts road users and their interactions as evolving nodes and edges in a spatiotemporal graph. Each node represents an individual road user, characterized by its state as features, such as location and velocity. A node-wise Long Short-Term Memory (LSTM) network is employed to capture the temporal evolution of these features. Edges are dynamically constructed based on spatial and temporal proximity, existing only when distance and time thresholds are met for modeling interaction relevance. The GAT learns attention-weighted representations of these dynamic interactions, which are subsequently used by a classifier to predict the risk of a crash. Experimental results demonstrate that the proposed GAT-based method achieves 86.1% prediction accuracy, highlighting its effectiveness for proactive collision risk assessment and its potential to inform real-time warning systems and preventive safety interventions. Full article

(This article belongs to the Special Issue Advances in Vehicle Dynamics and Road Safety: Technologies, Simulations and Applications, 2nd Edition)

►▼ Show Figures

Figure 1

Show export options Show export options

Select all

Export citation of selected articles as:

Error

Oops... you haven't selected anything for export.

Displaying article 1-50 on page 1 of 2000.

Go to page 1 2 3 4 5

Search Results (178,541)

Further Information

Guidelines

MDPI Initiatives

Follow MDPI