Search Results (197)

Foreign Object Debris (FOD) detection is paramount for airport operations. The precise identification and removal of FOD are critical for ensuring airplane flight safety. This study collected FOD images using optical imaging sensors installed at Urumqi Airport and created a custom FOD dataset based on these images. To address the challenges of small targets and complex backgrounds in the dataset, this paper proposes optimizations and improvements based on the advanced detection network Dab-Deformable. First, this paper introduces a Lightweight Deep-Shallow Feature Fusion algorithm (LDSFF), which integrates a hotspot sensing network and a spatial mapping enhancer aimed at focusing the model on significant regions. Second, we devise a Multi-Directional Deformable Channel Attention (MDDCA) module for rational feature weight allocation. Furthermore, a feedback mechanism is incorporated into the encoder structure, enhancing the model’s capacity to capture complex dependencies within sequential data. Additionally, when combined with a Threshold Selection (TS) algorithm, the model effectively mitigates the distraction caused by the serialization of multi-layer feature maps in the Transformer architecture. Experimental results on the optical small FOD dataset show that the proposed network achieves a robust performance and improved accuracy in FOD detection. Full article

We propose a hybrid Convolutional Graph Neural Network (C-GNN) for direction-of-arrival (DOA) estimation in sparse sensor arrays under low-snapshot conditions. The C-GNN architecture combines 1D convolutional layers for local spatial feature extraction with graph convolutional layers for global structural learning, effectively capturing both fine-grained and long-range array dependencies. Leveraging the difference coarray technique, the sparse array is transformed into a virtual uniform linear array (VULA) to enrich the spatial sampling; real-valued covariance matrices derived from the array measurements are used as the network’s input features. A final multi-layer perceptron (MLP) regression module then maps the learned representations to continuous DOA angle estimates. This approach capitalizes on the increased degrees of freedom offered by the virtual array while inherently incorporating the array’s geometric relationships via graph-based learning. The proposed C-GNN demonstrates robust performance in noisy, low-data scenarios, reliably estimating source angles even with very limited snapshots. By focusing on methodological innovation rather than bespoke architectural tuning, the framework shows promise for data-efficient DOA estimation in challenging practical conditions. Full article

Heat stress (HS) in dairy cattle triggers systemic physiological disruptions, including milk yield decline, immune suppression, and reproductive dysfunction, jeopardizing sustainable livestock production. While conventional single-omics or phenotypic studies have provided fragmented insights, they fail to decipher the multi-layered regulatory networks and gene–environment interactions underlying HS. This review integrates current knowledge on HS-induced physiological responses and molecular adaptations in dairy cattle from a multi-omics perspective, highlighting integrative approaches that combine IoT-enabled monitoring and AI-driven genetic improvement strategies. However, key challenges persist, such as complexities in bioinformatic data integration, CRISPR off-target effects, and ethical controversies. Future directions will emphasize the development and application of AI-aided predictive models to enable precision breeding, thereby advancing climate-resilient genetic improvement in dairy cattle. Full article

Electroencephalogram (EEG) signals provide a direct and non-invasive means of interpreting brain activity and are increasingly becoming valuable in embedded emotion-aware systems, particularly for applications in healthcare, wearable electronics, and human–machine interactions. Among various EEG-based emotion recognition techniques, deep learning methods have demonstrated superior performance compared to traditional approaches. This advantage stems from their ability to extract complex features—such as spectral–spatial connectivity, temporal dynamics, and non-linear patterns—from raw EEG data, leading to a more accurate and robust representation of emotional states and better adaptation to diverse data characteristics. This study explores and compares deep and shallow neural networks for human emotion recognition from raw EEG data, with the goal of enabling real-time processing in embedded and edge-deployable systems. Deep learning models—specifically convolutional neural networks (CNNs) and recurrent neural networks (RNNs)—have been benchmarked against traditional approaches such as the multi-layer perceptron (MLP), support vector machine (SVM), and k-nearest neighbors (kNN) algorithms. This comparative study investigates the effectiveness of deep learning techniques in EEG-based emotion recognition by classifying emotions into four categories based on the valence–arousal plane: high arousal, positive valence (HAPV); low arousal, positive valence (LAPV); high arousal, negative valence (HANV); and low arousal, negative valence (LANV). Evaluations were conducted using the DEAP dataset. The results indicate that both the CNN and RNN-STM models have a high classification performance in EEG-based emotion recognition, with an average accuracy of 90.13% and 93.36%, respectively, significantly outperforming shallow algorithms (MLP, SVM, kNN). Full article

The progressive proliferation of sensor deployment in distribution networks (DNs), propelled by the dual drivers of power automation and ubiquitous IoT infrastructure development, has precipitated exponential growth in real-time data generated by multisource heterogeneous (MSH) sensors within multilayer grid architectures. This phenomenon presents dual implications: large-scale datasets offer an enhanced foundation for reliability assessment and dispatch planning in DNs; the dramatic escalation in data volume imposes demands on the computational precision and response speed of traditional evaluation approaches. The identification of critical influencing factors under extreme operating conditions, coupled with dynamic assessment and prediction of DN reliability through MSH data approaches, has emerged as a pressing challenge to address. Through a brief analysis of existing technologies and algorithms, this article reviews the technological development of MSH data analysis in DNs. By integrating the stability advantages of conventional approaches in practice with the computational adaptability of artificial intelligence, this article focuses on discussing key approaches for MSH data processing and assessment. Based on the characteristics of DN data, e.g., diverse sources, heterogeneous structures, and complex correlations, this article proposes several practical future directions. It is expected to provide insights for practitioners in power systems and sensor data processing that offer technical inspirations for intelligent, reliable, and stable next-generation DN construction. Full article

The glass transition temperature (T_g) was a pivotal parameter governing the thermal and mechanical properties of bismaleimide-based polyimide (BMI) resins. However, limited experimental data for BMI systems posed significant challenges for predictive modeling. To address this gap, this study introduced a hybrid modeling framework leveraging transfer learning. Specifically, a multilayer perceptron (MLP) deep neural network was pre-trained on a large-scale polymer database and subsequently fine-tuned on a small-sample BMI dataset. Complementing this approach, six interpretable machine learning algorithms—random forest, ridge regression, k-nearest neighbors, Bayesian regression, support vector regression, and extreme gradient boosting—were employed to construct transparent predictive models. SHapley Additive exPlanations (SHAP) analysis was further utilized to quantify the relative contributions of molecular descriptors to T_g. Results demonstrated that the transfer learning strategy achieved superior predictive accuracy in data-scarce scenarios compared to direct training on the BMI dataset. SHAP analysis identified charge distribution inhomogeneity, molecular topology, and molecular surface area properties as the major influences on T_g. This integrated framework not only improved the prediction performance but also provided feasible insights into molecular structure design, laying a solid foundation for the rational engineering of high-performance BMI resins. Full article

Visual simultaneous localization and mapping (Visual SLAM) based on RGB-D image data includes two main tasks: One is to build an environment map, and the other is to simultaneously track the position and movement of visual odometry estimation (VOE). Visual SLAM and VOE are used in many applications, such as robot systems, autonomous mobile robots, assistance systems for the blind, human–machine interaction, industry, etc. To solve the computer vision problems in Visual SLAM and VOE from RGB-D images, deep learning (DL) is an approach that gives very convincing results. This manuscript examines the results, advantages, difficulties, and challenges of the problem of Visual SLAM and VOE based on DL. In this paper, the taxonomy is proposed to conduct a complete survey based on three methods to construct Visual SLAM and VOE from RGB-D images (1) using DL for the modules of the Visual SLAM and VOE systems; (2) using DL to supplement the modules of Visual SLAM and VOE systems; and (3) using end-to-end DL to build Visual SLAM and VOE systems. The 220 scientific publications on Visual SLAM, VOE, and related issues were surveyed. The studies were surveyed based on the order of methods, datasets, evaluation measures, and detailed results. In particular, studies on using DL to build Visual SLAM and VOE systems have analyzed the challenges, advantages, and disadvantages. We also proposed and published the TQU-SLAM benchmark dataset, and a comparative study on fine-tuning the VOE model using a Multi-Layer Fusion network (MLF-VO) framework was performed. The comparison results of VOE on the TQU-SLAM benchmark dataset range from 16.97 m to 57.61 m. This is a huge error compared to the VOE methods on the KITTI, TUM RGB-D SLAM, and ICL-NUIM datasets. Therefore, the dataset we publish is very challenging, especially in the opposite direction (OP-D) when collecting and annotation data. The results of the comparative study are also presented in detail and available. Full article

A millimeter-wave multi-layer and miniaturized multibeam antenna fed by an E-plane Butler matrix (BM) in substrate integrated waveguide (SIW) technology is proposed. For the beam-forming network (BFN), a folded E-plane 4 × 4 BM is proposed, whose basic components are stacked up along the vertical direction aiming to reduce the horizontal size by more than 75% compared with a single-layer BM. For the radiation portion, an unconventional slot antenna array arranged in a ladder type is adopted. The slot antenna elements are distributed in separate layers, making them more compatible with the presented BM and are arranged in the longitudinal direction to suppress the mutual coupling effect. Furthermore, the BM has been adjusted to accommodate the slot antenna array and obtain further miniaturization. The overall dimension of the designed multibeam antenna, taking the BFN into account, is 12 mm × 45 mm × 2 mm (1.2 λ × 4.5 λ × 0.2 λ), which is preferable for future 6G smartphone applications. The impacts of the air gap in fabrication are also taken into consideration to alleviate the error between simulated model and fabricated prototype. Full article

This review explores the intersection of Artificial Intelligence (AI) and Generative Design (GD) in engineering within the mechanical, industrial, civil, and architectural domains. Driven by advances in AI and computational resources, this intersection has grown rapidly, yielding over 14,000 publications since 2016. To map the research landscape, this review employed semantic search and Natural Language Processing, parsing 14,355 publications to ultimately select the 88 most relevant studies through clustering and topic modelling. These studies were categorised according to AI and GD techniques, application domains, benefits, and limitations, providing insights into research trends and practical implications. The results reveal a significant growth in the integration of advanced generative AI methods, notably Generative Adversarial Networks for direct design generation, alongside the continued use of genetic algorithms and surrogate models (e.g., Convolutional Neural Networks and Multilayer Perceptrons) to manage computational complexity. Structural and aerodynamic applications were the most common, with benefits including improvements in computational efficiency and design diversity. However, barriers remain, including data generation costs, model accuracy, and interpretability. Research opportunities include the development of generalisable foundation surrogate models, the integration of emerging generative methods such as diffusion models and large language models, and the explicit consideration of manufacturability constraints within generative processes. Full article

Several challenges hinder accurate and physically consistent dynamic parameter estimation in power systems, particularly under scenarios involving limited measurements, strong system nonlinearity, and high variability introduced by renewable integration. Although data-driven methods such as Physics-Informed Neural Networks (PINNs) provide a promising direction, they often suffer from poor generalization and training instability when faced with complex dynamic regimes. To address these challenges, we propose a Residual Physics-Informed Neural Network (Res-PINN) framework, which integrates a residual neural architecture with the swing equation to enhance estimation robustness and precision. By replacing the traditional multilayer perceptron (MLP) in PINN with residual connections and injecting normalized time into each network layer, the proposed model improves temporal awareness and enables stable training of deep networks. A physics-constrained loss formulation is employed to estimate inertia and damping parameters without relying on large-scale labeled datasets. Extensive experiments on a 4-bus, 2-generator power system demonstrate that Res-PINN achieves high parameter estimation accuracy across various dynamic conditions and outperforms traditional PINN and Unscented Kalman Filter (UKF) methods. It also exhibits strong robustness to noise and low sensitivity to hyperparameter variations. These results show the potential of Res-PINN to bridge the gap between physics-guided learning and practical power system modeling and parameter identification. Full article

Road transport is a major contributor to air pollution, necessitating sustainable solutions for urban logistics. This study presents a time-dependent vehicle routing problem (VRP) model aimed at minimizing fuel consumption and greenhouse gas emissions while addressing stochastic customer demands. By incorporating key environmental factors such as road gradients, vehicle load, temperature, wind direction, and asphalt type, the proposed model provides a comprehensive approach to reducing transportation-related pollutants. To solve the computationally complex problem, a hybrid algorithm combining the gray wolf optimizer (GWO) and the multilayer perceptron (MLP) neural network is introduced. The algorithm demonstrates superior performance, achieving an error rate of less than 2% for medium-scale problems and significantly reducing fuel and driver costs. Sensitivity analyses reveal the profound impact of environmental parameters, with wind speed and direction altering optimal routing in over 80% of cases for large-scale instances. This research advances green logistics by integrating dynamic environmental considerations into routing decisions, balancing economic objectives with sustainability. The proposed model and algorithm offer a scalable solution to real-world challenges, enabling policymakers and logistics planners to improve environmental outcomes while maintaining operational efficiency. Full article

The Fourth Industrial Revolution has introduced “shared manufacturing” as a key concept that leverages digitalization, IoT, blockchain, and robotics to redefine the production and delivery of manufacturing services. This paper presents a novel approach to decentralized warehouse management integrating Large Language Models (LLMs) into the decision-making processes of autonomous agents, which serves as a proof of concept for shared manufacturing. A multi-layered system architecture consisting of physical, digital shadow, organizational, and protocol layers was developed to enable seamless interactions between parcel and warehouse agents. Shared Warehouse game simulations were conducted to evaluate the performance of LLM-driven agents in managing warehouse services, including direct and pooled offers, in a competitive environment. The simulation results show that the LLM-controlled agent clearly outperformed traditional random strategies in decentralized warehouse management. In particular, it achieved higher warehouse utilization rates, more efficient resource allocation, and improved profitability in various competitive scenarios. The LLM agent consistently ensured optimal warehouse allocation and strategically selected offers, reducing empty capacity and maximizing revenue. In addition, the integration of LLMs improves the robustness of decision-making under uncertainty by mitigating the impact of randomness in the environment and ensuring consistent, contextualized responses. This work represents a significant advance in the application of AI to decentralized systems. It provides insights into the complexity of shared manufacturing networks and paves the way for future research in distributed production systems. Full article

This research aims to explore the interdisciplinary connection between the field of neurology and artificial intelligence (AI) through machine learning (ML) algorithms. The central objective is to evaluate the current state of research in the Neuro-ML field and identify gaps in the literature that require additional approaches. To achieve this objective, 10 analyses were introduced that analyze the distribution of articles based on keywords, countries, years, publishers, and ML algorithms used in the context of neurological diseases. Surveys were also conducted to identify the diseases most frequently studied through ML algorithms. Thus, it was found that Alzheimer’s disease (37 articles for Support Vector Regression—SVR; 31 for Random Forest—RF), Parkinson’s disease (46 articles for SVM and 48 for RF), and multiple sclerosis (9 articles for SVM) are the most studied diseases in the field of Neuro-ML. The study analyzes Alzheimer’s, Parkinson’s, and multiple sclerosis in detail by focusing on diagnosis. The overall results highlight an increase in researchers’ interest in applying ML in neurology, with models such as SVM (597 articles), Artificial Neural Network (525 articles), and RF (457 articles) being the most used. The results highlighted three major gaps: the underrepresentation of rare diseases, the lack of standardization in evaluating the performance of ML models, and the lack of exploration of algorithms with greater implementation difficulty, such as Extreme Gradient Boosting and Multilayer Perceptron. The value analysis of the performance metrics of ML models demonstrates the ability to correctly classify neuro-degenerative diseases, with high accuracy in some cases (for example, 97.46% accuracy in Alzheimer’s diagnosis), but there may still be improvements. Future directions include exploring rare diseases, investigating underutilized algorithms, and developing standardized protocols for evaluating the performance of ML models, which will facilitate the comparison of results across different studies. Full article

The neural moving horizon estimator (NMHE) is a relatively new and powerful state estimator that combines the strengths of neural networks (NNs) and model-based state estimation techniques. Various approaches exist for constructing NMHEs, each with its unique advantages and limitations. However, a comprehensive literature review that consolidates existing knowledge, outlines design guidelines, and highlights future research directions is currently lacking. To address this gap, this systematic review screened 1164 records and ultimately included 22 primary studies, following the preferred reporting items for systematic reviews and meta-analyses (PRISMA) protocol. This paper (1) explains the fundamental principles of NMHEs, (2) explores three major NMHE architectures, (3) analyzes the types of NNs used, such as multi-layer perceptrons (MLPs), long short-term memory networks (LSTMs), radial basis function networks (RBFs), and fuzzy neural networks, (4) reviews real-time implementability—including reported execution times ranging from 1.6 μs to 11.28 s on different computing hardware—and (5) identifies common limitations and future research directions. The findings show that NMHEs can be realized in three principal ways: model learning, cost function learning, and approximating the real-time optimization in moving horizon estimation. Cost function learning offers flexibility in capturing task-specific estimation goals, while model learning and optimization approximation approaches tend to improve estimation accuracy and computational speed, respectively. Full article

With the rapid expansion of digital communication and data sharing, ensuring robust security for sensitive information has become increasingly critical, particularly when data are transmitted over public networks. Traditional encryption techniques are increasingly vulnerable to evolving cyber threats, making single-layer security mechanisms less effective. This study proposes a multi-layered security approach that integrates cryptographic and steganographic techniques to enhance data protection. The framework leverages advanced methods such as encrypted data embedding in images, DNA sequence coding, QR codes, and least significant bit (LSB) steganography. To evaluate its effectiveness, experiments were conducted using text messages, text files, and images, with security assessments based on PSNR, MSE, SNR, and encryption–decryption times for text data. Image security was analyzed through visual inspection, correlation, entropy, standard deviation, key space analysis, randomness, and differential analysis. The proposed method demonstrated strong resilience against differential cryptanalysis, achieving high NPCR values (99.5784%, 99.4292%, and 99.5784%) and UACI values (33.5873%, 33.5149%, and 33.3745%), indicating robust diffusion and confusion properties. These results highlight the reliability and effectiveness of the proposed framework in safeguarding data integrity and confidentiality, providing a promising direction for future cryptographic research. Full article