Search Results (429)

The growing demand for high-fidelity, multi-parameter, distributed sensing in critical domains such as structural health monitoring, oil and gas exploration, and secure perimeter surveillance is pushing traditional optical fiber sensors (OFS) to their performance limits. Although conventional multiplexing techniques such as time-division and wavelength-division multiplexing (TDM, WDM) have been commercially successful, they are rapidly approaching fundamental bottlenecks in sensor density, spatial resolution, and data capacity. This review argues that the synergistic convergence of space-division multiplexing (SDM) and artificial intelligence (AI) represents a paradigm shift, enabling a new generation of intelligent, high-dimensional sensing networks. We comprehensively survey the state of the art in SDM-based OFS, detailing the operating principles and applications of multi-core fibers (MCFs) for ultra-dense sensor arrays and 3D shape sensing, as well as few-mode fibers (FMFs) for mode-division multiplexing and enhanced multi-parameter discrimination. However, the unprecedented spatial parallelism provided by SDM introduces significant challenges, including inter-channel crosstalk, complex signal demultiplexing, and massive data volumes. This paper systematically explores how AI, particularly machine learning (ML) and deep learning (DL), is being leveraged not merely as a tool but as an indispensable core technology to mitigate these impairments. We critically analyze AI’s role in digital crosstalk suppression, intelligent mode demultiplexing, signal denoising, and solving complex inverse problems for parameter estimation. Furthermore, we highlight how this AI–SDM synergy enables capabilities beyond the reach of either technology alone, such as super-resolution sensing and predictive analytics. The discussion is extended to include the critical supporting pillars of this ecosystem, such as advanced interrogation techniques and the associated data management challenges. Finally, we provide a forward-looking perspective on the trajectory of the field, outlining a path toward cognitive sensing networks that are self-calibrating, adaptive, and capable of autonomous decision-making. This review is intended to serve as a foundational reference for researchers and engineers at the intersection of photonics and intelligent systems, illuminating the pathway toward tomorrow’s intelligent sensing infrastructure. Full article

Underwater Acoustic Sensor Networks (UASNs) are pivotal for environmental monitoring, surveillance, and marine data collection. However, their open and largely unattended operational settings, constrained communication capabilities, limited energy resources, and susceptibility to insider attacks make it difficult to achieve safe, secure, and efficient collaborative learning. Federated learning (FL) offers a privacy-preserving method for decentralized model training but is inherently vulnerable to Byzantine threats and malicious participants. This paper proposes trust-aware FL for underwater sensor networks (TAFL-UWSN), a trust-aware FL framework designed to improve security, reliability, and energy efficiency in UASNs by incorporating trust evaluation directly into the FL process. The goal is to mitigate the impact of adversarial nodes while maintaining model performance in low-resource underwater environments. TAFL-UWSN integrates continuous trust scoring based on packet forwarding reliability, sensing consistency, and model deviation. Trust scores are used to weight or filter model updates both at the node level and the edge layer, where Autonomous Underwater Vehicles (AUVs) act as mobile aggregators. A trust-aware federated averaging algorithm is implemented, and extensive simulations are conducted in a custom Python-based environment, comparing TAFL-UWSN to standard FedAvg and Byzantine-resilient FL approaches under various attack conditions. TAFL-UWSN achieved a model accuracy exceeding 92% with up to 30% malicious nodes while maintaining a false positive rate below 5.5%. Communication overhead was reduced by 28%, and energy usage per node dropped by 33% compared to baseline methods. The TAFL-UWSN framework demonstrates that integrating trust into FL enables secure, efficient, and resilient underwater intelligence, validating its potential for broader application in distributed, resource-constrained environments. Full article

Reliable steering regulation under varying road curvature and actuator constraints remains a central challenge in autonomous electric vehicles (AEVs). Many exiting approaches rely on reactive error correction or treat preview information solely as a reference adjustment, limiting anticipation and physical consistency. This study proposes a prediction–preview steering control (PSC) framework in which future curvature information within state propagation and constraint handling enables forward-looking steering decisions while respecting dynamic and actuator limits. The method is evaluated using a lateral-heading vehicle model with real-road geometric variation. Experimental results indicate significant improvement in tracking performance, reducing lateral RMSE from 0.1747 m to 0.0074 m with a maximum deviation of 0.0889 m and limiting heading RMSE to 0.0867° (maximum 1.2046°). Steering angle commands remain bounded within ±8.7°, while steering angle rate is maintained within 40–60°/s, ensuring smooth and dynamically admissible operation. The proposed strategy offers a computationally efficient solution for embedded AEV steering systems and demonstrates improved robustness under practical curvature transitions. Full article

Detecting violent content in audio recordings is crucial for public safety, autonomous surveillance, and content moderation, particularly when visual cues are unreliable or unavailable. A resource-aware two-stage cascade system is proposed for acoustic violence detection that combines a lightweight Least Squares Linear Detector (LSLD) as a first-stage screener with a trimmed version of YAMNet as a second-stage classifier. A percentile-based forwarding rule controls the fraction of segments routed to the deep stage, turning the accuracy–cost trade-off into an explicit operating parameter for always-on deployment. The approach is evaluated on a publicly released dataset of real-world violent audio augmented with background noise and artificial reverberation. The results in the low-false-alarm regime show that the proposed cascade preserves performance close to a Stage 2-only baseline while substantially reducing average deep-inference workload. An ablation study validates the role of the LSLD as an inexpensive pre-filter, and robustness is assessed under clean, reverberant, and 12 dB noise conditions. Finally, an analytic energy consumption model is provided, which links computational workload to daily energy demand and photovoltaic sizing on ultra-low-power hardware, supporting sustainable off-grid deployment. Full article

Background: Hypermobile Ehlers–Danlos syndrome (hEDS) can present as a complex interplay of widespread symptomatology and multisystem involvement, posing diagnostic and treatment challenges. Objective characterization of cervical spine and neurovascular findings in hEDS has been limited. Previous studies have emphasized upper cervical spine complications in hEDS, yet the relevance and mechanisms underlying associated symptomatology have not been elucidated. This study examined objective test findings in patients with hEDS at an outpatient neck clinic to explore cervical spine and neurovascular pathology that could contribute to further understanding the clinical profile of a subset of patients with hEDS. Methods: This single-center, retrospective observational study included patients with hEDS aged 20–50 years from 1 January 2022–31 December 2024, at an outpatient neck center. It excluded previous neck surgery, traumatic events, or related injury. Demographic, clinical, and diagnostic data were collected through a retrospective chart review, including measurements from standard clinical diagnostic protocols: digital motion X-ray (videofluoroscopy), cone beam CT, Doppler ultrasound, and tonometry. Results: More than 71% of patients reported ≥29 symptoms. Nearly all patients exhibited co-occurring forward head, decreased depth of curve, ligamentous cervical instability, and decreased internal jugular vein (IJV) and vagus nerve cross-sectional area (CSA). Vagus nerve CSA was found to be significantly smaller than the comparative healthy/normal population. IJV CSA was significantly smaller at C1 than at C4–C5, suggesting evidence of carotid sheath compression at C1. Conclusions: This study offers novel evidence that cervical spine pathology, IJV compression, and vagus nerve degeneration are uniformly prevalent in hEDS, which may contribute to, or be an etiological basis for, the multisystem involvement in a subset of patients with this disorder. These findings provide hypothesis-generating data to inform future mechanistic and therapeutic studies, including exploration of new diagnostic and treatment targets. Full article

This paper addresses the multi-phase homing and docking missions of over-actuated autonomous underwater vehicles (AUVs), where switching among forward cruising, reverse braking, and hovering can induce actuator saturation, rate limit violations, and undesirable transients. We propose a unified framework that couples supervisory mode management with mode-driven constrained control allocation solved by a warm-started sequential quadratic programming (SQP) routine. The controllable wrench is modeled by a mode-dependent differentiable map constructed from the actuator models, and the allocator enforces amplitude bounds and per-cycle increment limits while trading off wrench tracking and actuator usage through mode-scheduled weights. To mitigate switching transients, a continuous transition factor is introduced to interpolate the desired wrench and dominant cost weights, and an integrator alignment reset is applied at switching instants to keep the outer-loop proportional–integral–derivative (PID) output continuous. The allocator is further warm-started by projecting the previous solution onto the post-switch constraint box. The framework is integrated into the Mission-Oriented Operating Suite–Interval Programming (MOOS-IvP) autonomy middleware with adaptive line-of-sight (ALOS) guidance and adaptive PID motion control and is validated on the TS-100 AUV in water tank experiments. Comparative results against a PID-only baseline without control allocation and a variant without bumpless switching show reduced roll transients during the reverse-to-hover transition and improved hover-mode depth station keeping while maintaining feasible actuator commands under constraints. Full article

Sweeteners occupy a pivotal role in the global transition toward sustainable, health-aligned, and resource-efficient food systems. Conventional sucrose production carries significant environmental burdens, while escalating metabolic health concerns intensify demand for viable alternatives. This paper reframes sweeteners not as commodity ingredients, but as digitally engineered, biologically manufactured, and circularity-optimized materials within the emerging bioeconomy. Advances in artificial intelligence (AI), metabolic engineering, precision fermentation, and lignocellulosic valorization are fundamentally reshaping sweetener innovation. We introduce the Sweetener Innovation 4.0 framework, in which AI functions as the integrative engine linking molecular design, bioprocess optimization, and system-level sustainability. Across diverse sweetener classes, including steviol glycosides, mogrosides, rare sugars, sweet proteins, and forestry-derived polyols, AI accelerates discovery, improves metabolic flux control, optimizes downstream processing and enables more adaptive manufacturing systems. This digital–biological convergence is progressively decoupling sweetness production from land-intensive agriculture, reducing dependence on geographically constrained crops, and enabling resilient, low-carbon manufacturing pathways. Comparative life-cycle assessments highlight substantial sustainability gains, but also reveal persistent methodological gaps, particularly in accounting for downstream-processing energy and digital infrastructure emissions. Socioeconomic analysis further underscores the importance of equitable transitions, transparent labeling, and effective consumer communication as fermentation-derived sweeteners enter global markets. Looking forward, we identify key frontiers for Sweetener Innovation 4.0, including de novo AI-designed sweeteners, autonomous fermentation systems, carbon-negative feedstocks, personalized sweetness modulation, and integrated circular biorefineries. Together, these developments position sweeteners as a top domain for demonstrating how AI, biotechnology, and sustainability principles can jointly reshape ingredient development and industrial systems within the 21st-century circular-economy. Full article

Wireless sensor networks (WSNs) are integral components of smart environments. These allow monitoring and communication to take place autonomously across distributed sensor nodes. Nevertheless, they suffer from constrained resources that make them susceptible to routine-layer attacks. These specifically involve blackhole, flooding, selective forwarding attack traffic and normal traffic. The conventional machine learning and deep learning methods employed are effective in catering to these attacks, yet they have generalization issues when the network conditions are dynamic. The models are generally trained on the local features that make them more dependable and less interpretable. To overcome these issues, this paper proposes an attention-driven transformer encoder for tabular WSN traffic, designed for robust and interpretable intrusion detection in WSNs. The model represents the WSN features as sequential tokens and employs multi-head self-attention to capture global and local dependencies among sensor attributes and employs a multi-head self-attention for capturing the local and global dependencies among the sensor attributes. The framework integrated several components, including normalization, chi-square-based feature selection, and positional embedding. These are followed by multi-layer transformer encoding blocks for the feature fusion and subsequent classification. The framework has been evaluated on the publicly available WSN dataset. Results have been shown to attain an accuracy of 99.37%, which makes it outperform the traditional deep learning baseline models. The comparative analysis has shown the model to be superior in terms of generalization and reduced convergence time. It further offers enhanced interpretability that makes it a good fit to be deployed in real-world scenarios where resources can be constrained. Full article

Confronted with the severe imperatives to food security posed by a growing population and the urgent need for sustainable development amid climate change, traditional agricultural models face significant resource-intensive efficiency bottlenecks. Deep learning-based image recognition is driving a future-oriented intelligent agricultural revolution by enabling high-throughput phenotyping and autonomous decision-making across the production chain. This paper systematically reviews key advancements in image recognition within modern agriculture, mapping the fundamental paradigm shift from traditional hand-crafted feature engineering to adaptive deep feature learning. We critically analyze technological implementation and performance across five core application scenarios: high-precision pest and disease diagnosis, spatio-temporal growth monitoring and yield prediction through multi-source image fusion, agricultural robots for automated harvesting, non-destructive quality inspection of products, and intelligent precision management of farmland. The review further identifies critical challenges hindering large-scale technology adoption, primarily centered on the high costs of constructing high-quality agricultural datasets and model robustness in complex field environments. Consequently, this study provides a comprehensive and forward-looking reference for advancing the deep integration of vision technology, thereby offering a strategic path toward achieving more intelligent, efficient, and sustainable global agricultural production systems in the digital era. Full article

In complex urban traffic scenes, reliable pedestrian trajectory prediction is essential for Automated and Connected Electric Vehicles (ACEVs) and active safety systems. Despite recent progress, many existing approaches still suffer from limited long-term prediction accuracy, redundant temporal features, and high computational cost, which restricts their deployment on vehicles with constrained onboard resources. To address these issues, this paper presents a lightweight trajectory prediction framework named BiTraP-DGF. The model adopts parallel Gated Recurrent Unit (GRU) and Long Short-Term Memory (LSTM) temporal encoders to extract motion information at different time scales, allowing both short-term motion changes and longer-term movement tendencies to be captured from observed trajectories. A conditional variational autoencoder (CVAE) with a bidirectional GRU decoder is further employed to model multimodal uncertainty, where forward prediction is combined with backward goal estimation to guide trajectory generation. In addition, a gated sparse attention mechanism is introduced to suppress irrelevant temporal responses and focus on informative time segments, thereby reducing unnecessary computation. Experimental results on the JAAD dataset show that BiTraP-DGF consistently outperforms the BiTraP-NP baseline. For a prediction horizon of 1.5 s, CADE is reduced by 20.9% and CFDE by 22.8%. These results indicate that the proposed framework achieves a practical balance between prediction accuracy and computational efficiency for autonomous driving applications. Full article

To tackle the navigation challenge in dynamic and complex environments, this study designs a fusion planning framework that synergistically integrates enhanced A* algorithm with improved DWA, inspired by the biological dual-layer navigation mechanism of global path planning and local real-time obstacle avoidance. Firstly, the original global path from the conventional A* algorithm is smoothed and length-reduced through a three-stage optimization strategy involving redundant node removal and forward and reverse path relaxation, mimicking the behavioral logic of honeybees and desert ants that eliminate redundant routes to complete foraging and homing with minimal energy consumption. Secondly, an evaluation function integrating dynamic obstacle perception and adaptive weight adjustment is designed for the DWA to enhance the intelligence of local planning, drawing on the adaptive strategy of animals such as antelopes that adjust behavioral priorities according to environmental complexity to balance safety and efficiency. To comprehensively verify the performance of the proposed algorithm, simulation evaluations are performed in various scenarios, including 20 × 20 and 30 × 30 grid maps, with single and dual dynamic obstacles. Results demonstrate that our algorithm outperforms conventional methods in path length, smoothness, and safety. Further physical verification is carried out on a LiDAR-equipped mobile robot (Shenzhen Yuanchuangxing Technology Co., Ltd., Shenzhen, China) based on the ROS platform, confirming that the algorithm can stably achieve static path tracking and real-time obstacle avoidance in real indoor environments. Consequently, the developed hybrid algorithm delivers a viable and robust solution for autonomous mobile robots to navigate safely and efficiently in unpredictable and complex environments. Full article

Large language models (LLMs) are fundamentally transforming intelligent traffic systems by enabling semantic abstraction, probabilistic reasoning, and multimodal information fusion across heterogeneous data. This review examines existing research on LLM integration, ranging from data representation to autonomous agents, through an information-theoretic lens, conceptualizing LLMs as entropy-minimizing probabilistic systems that shape their capabilities in uncertainty modeling and semantic compression. We identify core integration patterns and analyze fundamental limitations arising from the inherent mismatch between discrete, entropy-driven LLM reasoning and the continuous, causal, and safety-critical nature of physical traffic environments. This reflects a deep structural tension rather than mere technical gaps. We delineate clear boundaries: LLMs are indispensable for managing high semantic entropy in tasks like contextual understanding and knowledge integration, whereas classical physics-based and optimization models remain essential in domains requiring ultra-low physical, temporal, and causal/normative entropy, such as real-time control and safety verification. Finally, we propose a forward-looking research agenda centered on hybrid intelligence architectures that bridge semantic information processing with physical system modeling for next-generation traffic systems. Full article

Industry 4.0 envisions a highly interconnected, autonomous manufacturing ecosystem enabled by the Industrial Internet of Things, Cyber-Physical Systems, and Artificial Intelligence. The emergence of large language models introduces new capabilities for semantic-aware decision-making, cross-domain knowledge integration, and intelligent automation. However, privacy, security, and regulatory constraints often isolate industrial data, impeding the scalability of LLMs in manufacturing. Federated learning addresses this by enabling decentralized LLM optimization without exposing raw data. This paper presents a comprehensive review of recent federated large language model research with a focus on industrial feasibility, comparing enabling techniques, system designs, and deployment strategies. Based on existing studies, forward-looking analyses are provided to highlight potential challenges and trade-offs in practical adoption, including computation and communication overheads, synchronization in large-scale federations, and system robustness. By bridging foundational methods with emerging industrial scenarios, we finally discuss the significant challenges associated with deploying federated large language models in complex industrial environments and outline a future research agenda. Full article

Supercapacitor modules for energy storage systems often require complex active balancing circuits to manage voltage imbalances between series-connected cells. This paper proposes a modified forward converter topology that passively charges and balances supercapacitor modules simultaneously. The proposed solution is modular, provides galvanic isolation, and is self-regulating, eliminating the need for dedicated sensors or complex control logic. Voltage equalization is achieved autonomously through coupled inductors, naturally directing current to the cells with the lowest voltage during the period when the converter is off. This work details the operating principle of the converter and analyzes two architectures: a non-crossover configuration and a crossover configuration. This study validated the system performance through PSIM simulations and a hardware prototype. The experimental results demonstrate that both configurations successfully charge and balance the supercapacitors. However, the crossover and non-crossover configurations achieve faster equalization under certain imbalance conditions. In contrast, the crossed configuration exhibits a smaller final voltage discrepancy between cells compared to the non-crossover architecture. The proposed converter proves to be a simple, robust, and effective solution for managing supercapacitor modules. Full article

Accurate braking-distance prediction for heavy-duty multi-axle trucks remains challenging due to the large gross vehicle weight, tandem-axle interactions, and strong transient load transfer during emergency braking. Recent studies on tire–road friction estimation, commercial-vehicle braking control (EBS/AEBS), and weigh-in-motion (WIM) sensing have highlighted that unmeasured vertical-load dynamics and time-varying friction are key sources of prediction uncertainty. To address these limitations, this study proposes an integrated sensing–simulation–AI framework that combines real-time axle-load estimation, full-scale robotic braking tests, fused road-friction sensing, and physics-consistent machine-learning modeling. A micro-electro-mechanical systems (MEMS)-based load-angle sensor was installed on the leaf-spring panel linking tandem axles, enabling the continuous estimation of dynamic vertical loads via a polynomial calibration model. Full-scale on-road braking tests were conducted at 40–60 km/h under systematically varied payloads (0–15.5 t) using an actuator-based braking robot to eliminate driver variability. A forward-looking optical friction module was synchronized with dynamic axle-load estimates and deceleration signals, and additional scenarios generated in a commercial ASM environment expanded the operational domain across a broader range of friction, grade, and loading conditions. A gradient-boosting regression model trained on the hybrid dataset reproduced measured stopping distances with a mean absolute error (MAE) of 1.58 m and a mean absolute percentage error (MAPE) of 2.46%, with most predictions falling within ±5 m across all test conditions. The results indicate that incorporating real-time dynamic axle-load sensing together with fused friction estimation improves braking-distance prediction compared with static-load assumptions and purely kinematic formulations. The proposed load-aware framework provides a scalable basis for advanced driver-assistance functions, autonomous emergency braking for heavy trucks, and infrastructure-integrated freight safety management. All full-scale braking tests were carried out at approximately 60% of the nominal service-brake pressure, representing non-panic but moderately severe braking conditions, and the proposed model is designed to accurately predict the resulting stopping distance under this prescribed braking regime rather than to minimize the absolute stopping distance itself. Full article