Search Results (6,218)

Weather forecasting, given the ever-increasing occurrence of climate change-induced events, has been widely introduced as a method to offer accurate and timely forecasts for proactive measures and risk mitigation. Artificial intelligence of things (AIoT) offers promising solutions for short-term weather forecasting, contributing to the advancement of sustainable and efficient weather monitoring technologies. This work presents everWeather_2.0, a significantly enhanced low-cost and self-powered AIoT-based weather forecasting station, which addresses key challenges in power consumption, user engagement and forecasting accuracy. The proposed end-to-end Cloud-Edge-IoT (CEI) proof-of-concept solution improves upon its predecessor by combining a more robust renewable energy subsystem for complete power autonomy with a series of lightweight, adaptive statistical models for on-device forecasting and an integrated display for on-site user engagement. Deployed in a real-world scenario, the station demonstrated seamless operation and high short-term forecasting accuracy for the thermodynamic variables during the pilot deployment period, with model errors observed as low as 2% for 30 min forecasts to 4.3% for 120 min intervals, validating its applicability in real-time and continuous physical weather monitoring. While wind speed and rainfall were monitored, they were excluded from the current accuracy metrics due to their high volatility and the insufficient number of events recorded during the pilot period to ensure reliable modeling. Full article

With the rapid expansion of rail transit networks and increasing operational density, foreign object intrusion on tracks has emerged as a critical threat to train safety. Conventional manual inspection methods suffer from low efficiency, high miss rates, and inadequate real-time performance, failing to meet the stringent requirements of modern intelligent railway maintenance. While deep learning offers a promising paradigm shift, existing models often struggle with complex background interference and multi-scale target detection in railway scenarios. To address these challenges, this paper proposes CMF-Net, a unified detection framework for railway track foreign object detection. The CGG module serves as a lightweight feature extraction unit in the backbone, mitigating gradient vanishing and overfitting. The MSAF module enables adaptive multi-scale feature fusion via dual attention (CBAM), enhancing small-object detectability. The FGAF module captures fine-grained edges and textures through a four-branch decomposed convolution and fine-grained attention, suppressing complex background interference. The BiFPN module restructures the neck for efficient bidirectional cross-scale feature fusion. Furthermore, the TPSA module injects explicit railway-domain prior knowledge by fusing a learnable rail-centerline distance-decay field with the CBAM spatial attention map, guiding the detector to focus on operational danger zones and reducing false positives. Experiments on the OFBDs dataset demonstrate that CMF-Net achieves a mean Average Precision (mAP50) of 89.2% and an mAP50:95 of 64.5%, surpassing the baseline YOLOv5s by 4.8 pp and 5.3 pp, respectively. The model maintains a compact parameter size of 5.4 M, a computational cost of 15.2 GFLOPs, and real-time inference capability (56.2 FPS). Edge-deployment feasibility is validated via on-device benchmarking on three Jetson platforms (Nano, Xavier NX, and Orin Nano), where INT8 TensorRT inference achieves 16.2, 108.7, and 153.8 FPS, respectively, under one-hour continuous-inference soak tests with peak power below 16 W and steady-state junction temperatures within safe thermal margins. Statistical significance testing (p < 0.05) confirms the stability of these performance gains. These results indicate that CMF-Net provides rapid and accurate detection of various track intrusions, enabling robust real-time monitoring in dynamic railway environments and enhancing operational safety and intelligence. Full article

This study aims to develop a sustainable, low-cost, and high-performance supercapacitor electrode by valorizing waste date seeds (Phoenix dactylifera) into activated carbon and integrating it with a polymer-based hydrogel electrolyte. Waste date seeds were successfully converted into high-performance activated carbon through hydrothermal carbonization followed by sulfuric acid (H₂SO₄) chemical activation. The obtained date seed activated carbon (DSAC) was applied as an electrode material and incorporated into a hydrogel electrolyte for supercapacitor applications. Structural, thermal, and morphological analyses using SEM, FTIR, XRD, and TGA confirmed the formation of a predominantly microporous carbon framework enriched with oxygen-containing functional groups, indicating effective carbonization and activation. The porous structure and surface chemistry contributed to enhanced electrochemical behavior. The electrochemical behavior of the prepared DSAC electrode was investigated through cyclic voltammetry (CV) and galvanostatic charge–discharge (GCD) analyses. The material exhibited a highest specific capacitance of 179 F g⁻¹ at a scan rate of 5 mV s⁻¹ and 159 F g⁻¹ at a current density of 0.2 A g⁻¹, demonstrating reliable and stable capacitive characteristics suitable for biomass-derived carbon-based supercapacitor applications. The device also exhibited excellent cycling stability over 5500 cycles, confirming long-term durability. The results demonstrate a promising and environmentally friendly strategy for advanced energy storage systems. Furthermore, the sustainability and cost-effectiveness of the proposed approach are attributed to the utilization of abundant date seed biomass and the simplicity of the hydrothermal–chemical activation process. The enhanced electrochemical performance is primarily associated with the hierarchical porous structure of the activated carbon and the improved ion transport facilitated by the hydrogel electrolyte, which collectively contribute to stable capacitive behavior and long-term cycling durability. Full article

To address challenges such as severe occlusion caused by the dense growth of blueberry fruits in natural environments, complex backgrounds, and the limited computational resources of agricultural edge devices, this study proposes BR-DETR-Prune, a lightweight object detection model oriented towards edge computing environments. Based on the RT-DETR architecture, the model introduces a PConv-based FasterNet as the backbone network, which effectively reduces memory access latency and floating-point operation costs. Furthermore, it utilizes a “Gather-and-Distribute” (GD) mechanism to reconstruct the feature fusion neck. Through the unified aggregation and multi-branch distribution of global information, it significantly enhances the model’s feature extraction capability for dense and overlapping targets. An AIFI-RepBN encoder is designed, integrating re-parameterization technology into the attention module to further reduce computational redundancy. For lightweight processing, a random channel pruning strategy based on the “Lottery Ticket Hypothesis” is adopted to perform structural compression and fine-tuning on the model, achieving a significant reduction in the number of parameters while inversely improving accuracy. The experimental results demonstrate that BR-DETR-Prune achieves an mAP@0.5 of 97.1% on a self-built blueberry dataset, with only 15.52 M parameters and a computational load reduced to 34.0 GFLOPs. Its comprehensive performance is superior to mainstream models such as YOLOv8, YOLO11, and the original RT-DETR. Particularly, deployment testing on the NVIDIA Jetson Orin Nano Super embedded edge computing platform reveals that the model achieves a real-time inference speed of 20.5 FPS under FP16 precision, exhibiting smooth detection frames and strong robustness against occlusion. This study provides an effective optimization solution for the deployment of high-precision Transformer architectures on low-computational-power devices, offering an efficient and reliable visual perception approach for automated blueberry harvesting and yield estimation. Full article

Currently, there is a widespread use of inertial motion units (IMUs) based on micromechanical systems (MEMS) with applications ranging from consumer electronics to medical devices. One of the main uses of this technology is in human body motion capture systems, which require attaching various IMUs to the body. It is customary to start the design of IMU-based motion capture solutions by using generic off-the-shelf (OTS) solutions or custom integrations. However, it is common that generic OTS solutions or custom IMUs integrations are not necessarily intended or designed to be directly attached to the human body. To address this issue, a widely adopted solution is to perform quick workarounds to enable the IMUs to be “worn” by prospective users. However, this can have the drawbacks of increased probability of detachment, improper fit, user discomfort, adding noise to the IMU measurements, etc. Therefore, the development of OTS IMU-based motion capture solutions would greatly benefit from having a methodological approach for the design of device housings and/or adaptations for OTS solutions or custom IMU integrations, such that they can be effectively used as wearable devices. In this work, we introduce a design methodology for wearable devices based on the Theory of Inventive Problem Solving (TRIZ). By designing a “wearable device housing” for an OTS IMU solution, we show that the proposed TRIZ-based methodology provides a straightforward and structured approach for the design of wearable devices. Furthermore, we illustrate how various challenges encountered in the early stages of prototype development can be effectively addressed using this methodology. The results obtained with the study case confirm that the proposed TRIZ-based methodology effectively overcomes the challenges associated with the design of wearable devices based on generic OTS solutions or custom IMU integrations. Full article

The use of smartphones as analytical instruments is becoming increasingly widespread due to their ease of use and low cost. However, it has limitations, such as dependence on the smartphone’s sensor, the light source and the environment, which hinders the reproducibility and comparability of results. This paper presents the development of a portable device, called DUNI, which can be attached to any smartphone and is designed to overcome these limitations. The device, manufactured using 3D printing and with an average cost of €35, consists of an integrating sphere, which incorporates a lighting-electronic system, as well as accessories for measuring on different surfaces. It has been optimised by evaluating the influence of the optical geometry, the size and reflective coating of the sphere, the lighting conditions, and the electronic stability on measurement performance. It has been applied to the determination of hydrogen peroxide and biogenic amines in synthetic samples, achieving relative errors of less than 5% and detection limits between 3 and 6 µM. Overall, the device we have developed constitutes a portable, versatile and low-cost platform that enables quantitative colorimetric measurements using smartphones under controlled lighting conditions, with potential applications in on-site analysis and resource-limited settings. Full article

Rapid, accurate, and scalable ion sensing technologies are highly desirable for future flexible healthcare and lab-on-a-chip applications. Here, we present a fully roll-to-roll (R2R) gravure-printed single-walled carbon nanotube complementary ring oscillator (SWCNT-cRO)-based microfluidic ion sensing platform fabricated on a flexible substrate. The proposed platform combines scalable printed complementary electronics with frequency-based ion sensing via electrostatically induced top-gating in aqueous microfluidic environments. The fabricated SWCNT-cRO devices exhibited stable oscillation characteristics, with a high device yield (>80%) and continuous manufacturing capability at a web speed of 5.4 m/min. Printable ethanolamine/zirconium acetylacetonate-based n-doping technology enabled complementary SWCNT transistor operation, while multilayer CYTOP/FG-3650 encapsulation ensured stable electrical operation under ionic aqueous conditions. After integration into a polydimethylsiloxane-based microfluidic channel, the oscillation frequency of the SWCNT-cRO was systematically modulated by Na⁺ concentration and pH. The sensing mechanism was based on electrostatically induced carrier modulation in n-type SWCNT transistors, resulting in variations in propagation delay and corresponding shifts in oscillation frequency. Compared with conventional ion-sensitive transistor platforms, the proposed approach offers scalable manufacturing, non-contact ion sensing, elimination of external reference electrodes, and direct compatibility with digital frequency-signal processing systems. This work establishes a promising strategy for future low-cost, disposable, and flexible microfluidic sensing platforms for wearable healthcare and lab-on-a-chip applications, ion sensing, and thin-film transistors. Full article

The optimisation design of agricultural machinery is shifting from offline, experience-driven engineering towards adaptive, data-driven, and closed-loop intelligent optimisation. Conventional approaches based on computer-aided engineering (CAE), empirical testing, mathematical modelling, and static multi-objective optimisation have provided an important engineering foundation, but they remain limited under unstructured field conditions involving soil heterogeneity, crop variability, climatic disturbance, and nonlinear machinery–environment interactions. This review systematically examines the evolution of intelligent optimisation design for agricultural machinery from conventional simulation-based methods to artificial intelligence (AI)- and digital twin (DT)-enabled paradigms. First, mathematical modelling, response surface methodology, discrete element method (DEM), computational fluid dynamics (CFD), multi-body dynamics (MBD), heuristic algorithms, and early AI-assisted surrogate optimisation are reviewed to clarify their contributions and limitations. Second, frontier enabling technologies are analysed, including agriculture-specific large models, generative AI, lightweight edge intelligence, deep reinforcement learning (DRL), embodied AI, federated learning (FL), and privacy-preserving computing. Third, system-level applications integrating DT and AI are discussed, with emphasis on full-lifecycle machinery optimisation, device–edge–cloud collaborative control, multi-agent fleet coordination, predictive maintenance, and Agriculture 5.0-oriented intelligent equipment systems. Key deployment bottlenecks are further identified, including sim-to-real inconsistency, virtual–physical mismatch in DTs, edge-side trade-offs among accuracy, latency, energy consumption, and cost, insufficient validation standards, and economic adoption barriers. Finally, a 2025–2030 roadmap is proposed, highlighting large-model–DT closed loops, control biomimetics, green low-carbon optimisation, and trustworthy human–machine symbiosis for sustainable Agriculture 5.0. Full article

Wi-Fi security analysis and testing tools are vital to ensure the safety of wireless networks. Specialised hardware and software are needed to examine the underlying technology that connects our devices wirelessly. This article explores the feasibility of utilising the ESP32-S3 microcontroller as the basis for a low-cost, open-source, portable Wi-Fi penetration testing tool. By developing and evaluating the Deauther32 firmware, the project demonstrates key functionalities such as capturing and injecting frames to execute common Wi-Fi attacks, like beacon flooding and deauthentication. The developed HackHeld32 design complements the firmware by offering a compact and extendable handheld device, making the tool standalone and portable. These prototypes build upon previous work, the ESP8266 Deauther and the HackHeld Vega, by introducing significant improvements in functionality, usability, and hardware capabilities. This establishes a strong foundation for future development by demonstrating the potential of microcontroller-based solutions. These tools bridge the gap between accessibility for beginners and functionality for professionals by offering a cost-effective and portable solution for Wi-Fi security testing and beyond. Full article

The rapid evolution of next-generation electronics urgently demands high-performance functional materials. Two-dimensional (2D) semiconductors, characterized by tunable bandgaps, magnetic properties, and excellent optical and electronic properties, hold significant potential for applications in nanoelectronic devices, magnetic storage, and optoelectronics. However, the high computational cost of traditional Density Functional Theory (DFT) severely restricts large-scale high-throughput screening. Meanwhile, problems such as insufficient datasets and non-uniform data quality remain prevalent. Against this background, machine learning (ML), which captures intricate nonlinear correlations and accelerates the discovery of novel materials, has emerged as an efficient technical approach. This review systematically summarizes recent advances in ML-driven property prediction for 2D semiconductors. It first elaborates the fundamental properties and classifications of 2D semiconductors, and then compares traditional computational simulations with ML algorithms, clarifying the distinct advantages of data-driven approaches. Subsequently, this work focuses on the latest progress in predicting critical properties, including bandgap, magnetism, and other physical characteristics. For bandgap prediction, classical algorithms such as random forests are compared with deep learning models represented by graph neural networks. The results demonstrate that deep learning performs much better in low-data regimes and complex material systems. For magnetic property prediction, the impact of feature engineering strategies on model accuracy and efficiency is systematically analyzed. In addition, the research progress of other physical property prediction tasks is briefly summarized. Finally, future research directions for machine learning, including standardized materials databases, physics-informed machine learning, multimodal modeling, and the integration of machine learning with experimental and theoretical methods, are outlined to address challenges in data quality, model interpretability, and cross-system generalization ability. This work aims to provide a systematic theoretical foundation and methodological guidance for research on two-dimensional semiconductor materials assisted by machine learning. Full article

Detection of leaks in hydrogen (H₂) infrastructure is required on a large scale to enable a safe widespread use of this clean energy source. Sensing solutions must be low-cost, use scalable fabrication methods and allow multiplexed detection while providing reliable safety alarms as fast as possible. Optical methods can make this possible while avoiding the risk of ignition due to electronics at the point of detection. Metal hydride-based micro-mirror configurations benefit from a simple interrogation scheme, as long as the sensitive element can produce a large optical response. Magnesium thin films undergo a drastic variation of properties when hydrogenated, making them suitable for this application. In this work, a micro-mirror device using single-mode fibers capable of detecting the presence of H₂ with a loading t₁₀ and t₉₀ of 1.2 and 3.0 s, respectively, is demonstrated. A complete interrogation unit was developed, presenting a solution suited for widespread deployment using industry-standard optical components and equipment. Full article

In the fields of cell engineering, bio-fabrication, and targeted therapy, achieving high-precision manipulation of microparticles and cells remains a technical challenge. Although acoustic tweezers based on surface acoustic waves (SAWs) offer a promising solution, the structural complexity of conventional SAW devices has limited their practical applications. This work proposes a symmetric interdigitated transducer (IDT)-based acoustic tweezers device featuring a simple structure and high flexibility for modulating acoustic pressure field patterns and enabling particle manipulation. Theoretical investigations into the particle manipulation mechanism of the proposed device were conducted using the finite element method. A detachable polymethyl methacrylate (PMMA) assembly chamber was also designed. The effectiveness of the device was validated through dynamic and reconfigurable manipulation experiments using fluorescent polystyrene microspheres. Experimental results demonstrate that the proposed device can rapidly and precisely modulate SAW to achieve array-based manipulation of particle clusters, forming corresponding array patterns. Compared with conventional sorting methods, this device offers advantages including low cost, high precision, ease of operation, and good biocompatibility, making it suitable for large-scale manipulation of microparticles and biological cells. This technology has the potential to expand the application landscape of SAW and may emerge as a cutting-edge approach for directed cell assembly and culture. Full article

In this paper, a distributed quasi-generalized Reed–Solomon (Q-GRS)-coded cooperative split labeling diversity (DQ-GRSCC-SLD) scheme is proposed to support reliable cooperative transmission of small-volume information in typical scenarios such as device-to-device (D2D) communication, vehicular ad hoc networks (VANETs) and wireless sensor networks. The system employs distinct labeling mappers at the source and the relay, enabling single-antenna transmission while constructing equivalently a dual-antenna labeling diversity model at the destination, which enhances interference resistance and reduces transmission costs. In addition, an ingenious design is proposed to ensure that the destination obtains the joint Q-GRS code. To optimize the weight distribution of the joint code, a traversal search (TS) algorithm is developed. Furthermore, a low-complexity joint decoding algorithm for Q-GRS codes, namely bracketing decoding, is presented by leveraging the efficient decoding algorithm of generalized Reed–Solomon (GRS) codes. Compared to the conventional maximum likelihood (ML) decoding, its complexity has been reduced from comparing $q^k$ codewords to evaluating q or $q+1$ promising codewords. A theoretical performance analysis of the DQ-GRSCC-SLD scheme is provided. Simulation results reveal that the proposed DQ-GRSCC-SLD scheme demonstrates its superior performance under practical scenarios. Full article

Ultra-low-expansion (ULE) glass serves as a critical material in high-precision optical devices and semiconductor manufacturing; however, its inherent hardness and brittleness pose significant challenges for machining processes. During the diamond cutting of ULE glass, severe tool wear emerges as the primary factor limiting machined quality, which not only shortens tool life but also prolongs subsequent polishing time, thereby increasing processing costs and hindering sustainable manufacturing. To address this challenge, in situ laser assisted diamond cutting (LADC) has emerged as a promising technique for the sustainable machining of difficult-to-machine materials. In this study, for achieving sustainable machining of ULE glass, the effects of cutting speed on surface roughness and tool wear were systematically investigated. To determine the optimal parameter combination for minimizing surface roughness and tool wear simultaneously, an integrated optimization approach combining artificial neural network (ANN) and non-dominated sorting genetic algorithm II (NSGA-II) was employed. The experimental results indicated that a spindle speed of 2900 rpm and a feed speed of 1.1 mm/min was ascertained as the optimum combination to attain the desired outcomes for in situ LADC of ULE glass. Under the optimum machining parameters, in situ LADC resulted in a 70.08% reduction in surface roughness and 61.24% reduction in tool wear compared to conventional diamond cutting (CDC). This study demonstrates that in situ LADC can be recognized as a promising sustainable machining technique for machining of ULE glass. Full article

Monitoring eccentric hamstring strength is critical for reducing injury risk in soccer players, yet laboratory-based technologies such as isokinetic dynamometry remain costly and impractical for field use. The purpose of this study was to examine the association and exploratory predictive relationship between variables derived from a smartphone application (My Jump Lab) and eccentric hamstring strength outputs obtained with an instrumented field device (NordBord, Vald Performance, Australia), while also quantifying their absolute agreement during the Nordic hamstring exercise (NHE). Thirty-one professional soccer players from a second-division United Arab Emirates team performed the NHE on the NordBord, while a simultaneous two-dimensional (2D) kinematic analysis was conducted using the My Jump Lab app (version 5.0 for iOS; My Jump Lab, Madrid, Spain). Pearson correlations, linear regression models, and Bland–Altman analyses were used to distinguish linear association/prediction from agreement/interchangeability. Results revealed a very large association between My Jump Lab-derived torque estimates and NordBord peak torque (r = 0.77, p < 0.001), with moderate associations for breakpoint angle (r = 0.42–0.43). A combined regression model using My Jump Lab torque and breakpoint angle explained 69.2% of the variance in NordBord torque (SEE = 15.30 N·m), although this predictive result should be interpreted as exploratory because the variables are task-specific and partly share anthropometric and mechanical determinants. Bland–Altman analysis revealed poor agreement, with a large systematic difference and proportional bias, indicating that My Jump Lab overestimated torque values at higher strength levels (mean bias = +511.9 N·m). Therefore, torque values derived from the app should be interpreted as relative indicators rather than absolute equivalents to instrumented measurements. From a practical perspective, My Jump Lab may offer a low-cost option for broad screening or relative group profiling when instrumented devices are unavailable, but it should not be used as a substitute for instrumented devices or for individual longitudinal monitoring based on absolute torque values. Full article