Search Results (7,566)

Translating large-scale motion datasets into robust, deployable humanoid controllers is a critical challenge in engineering informatics, primarily due to the scarcity of high-quality annotations, the risk of mode collapse in conditional generation, and the strict constraints of onboard computing hardware. This paper presents a deployable two-stage learning system that maps clip-level motion datasets to a single-policy multi-skill controller and its deployable counterpart. We adopt coarse one-hot skill labels that can be assigned automatically at the clip level with negligible manual effort, enabling scalable dataset construction. To prevent conditional discriminators from ignoring skill conditions, we inject mismatched (transition, label) pairs and introduce a condition-aware loss that explicitly penalizes incorrect transition–label associations, improving controllability and mitigating mode collapse. For real-world deployment, we further propose a two-stage training strategy: a privileged teacher policy is first trained in simulation and then distilled into a student policy that relies on stacked historical proprioceptive observations, ensuring robustness against sensing noise and latency without relying on external state estimation. Extensive evaluations in simulation and on real hardware demonstrate improved skill coverage, transition coverage, realism, and training efficiency across heterogeneous embodiments. With the onboard computer of a Unitree G1 robot, the distilled policy runs at 100 Hz with 15–25 ms latency, confirming the system’s engineering feasibility. Full article

We developed an augmented Sungka board that integrates traditional Filipino gameplay with embedded sensor technology. Each pit is equipped with load cell sensors and HX711 analog-to-digital converters to accurately detect marble distribution and movement in real time. A Raspberry Pi 4 serves as the central controller, handling sensor data acquisition, game state processing, rule enforcement, and output display through a liquid crystal display. The system enables automatic score tracking, move validation, and real-time board updates without altering the physical structure or rules of Sungka. A rule-based decision algorithm using fuzzy logic and heuristic search evaluates possible moves in constant time, allowing seamless real-time interaction. Across 10,000 simulated games, the algorithm achieved win rates of 84.9% against random, 77.7% against greedy, and 56.3% against exact-match strategies, with statistically consistent performance. By combining reliable hardware sensing with intelligent decision support, the proposed system enhances engagement while preserving the cultural authenticity of Sungka. Full article

In the context of highly interconnected global markets and the rapid dissemination of multilingual information, traditional risk prediction methods that rely on single numerical sequences or monolingual text are insufficient for achieving early perception of cross-market risks. To address this issue, a cross-market risk early warning framework based on multilingual large language models and multimodal sensing fusion is proposed. The proposed approach is centered on a unified risk semantic space, where cross-lingual semantic alignment is employed to reduce semantic discrepancies across languages. Furthermore, a semantic–volatility coupling attention mechanism is introduced to capture the dynamic relationship between textual semantic evolution and market fluctuations. In addition, cross-market knowledge transfer and low-resource enhancement strategies are incorporated to improve the model’s generalization capability across multilingual and multi-market environments, thereby establishing an intelligent perception and early warning system for complex sensing scenarios. Experimental results demonstrate that the proposed method significantly outperforms multiple baseline models in multilingual cross-market risk prediction tasks. In the main experiment, the model achieves a root mean squared error (RMSE) of $0.1127$, an mean absolute error (MAE) of $0.0846$, and an area under the curve (AUC) of $0.8879$, while the early warning gain is improved to $5.2$ days, which is substantially better than the Transformer model (RMSE $0.1365$, AUC $0.8042$) and the multilingual BERT-based fusion model (AUC $0.8395$). In terms of classification performance, higher accuracy, precision, and recall are consistently achieved, with overall accuracy exceeding $0.88$, and both precision and recall are maintained above $0.85$, indicating strong discriminative capability in risk identification tasks. Cross-lingual generalization experiments further verify the robustness of the proposed framework. When trained solely on the English market, the model achieves AUC values of $0.8624$ and $0.8471$ on the Chinese and European markets, respectively, with RMSE reduced to $0.1185$, significantly outperforming competing methods. Overall, the proposed approach achieves substantial improvements in prediction accuracy, cross-lingual generalization, and early warning performance, providing an effective solution for artificial intelligence-driven sensing and risk early warning. Full article

During the last 15 years, the territorial strategy of the Common Fisheries Policy (CFP) has supported initiatives focused on promoting the sustainable growth of European fishing communities, such as establishing Museums of the Sea. These museums emphasize the preservation, safeguarding, and enhancement of both tangible and intangible maritime cultural heritage, turning territorial and identity resources into valuable assets with significant potential for cultural and educational tourism. They are essential in enhancing local identity and sense of belonging, along with the social appreciation of the fishing profession. This research collects and examines data originating from five Museums of the Sea founded since 2000 in the province of A Coruña (Galicia, Spain) with CFP financing. Findings emphasize the connections between the Museums of the Sea, education and tourism, creating opportunities for local growth in fishing-reliant areas, promoting economic variety, safeguarding maritime heritage, and strengthening maritime identity. Full article

Sleep plays a fundamental role in human health and cognitive functioning, motivating the development of reliable and scalable methodologies for sleep stage classification (SSC). Recent advances in non-invasive and economically sustainable sensing technologies enable continuous sleep monitoring beyond laboratory settings. However, SSC remains a challenging data analytics task due to the intrinsic class imbalance among sleep stages. This study investigates the effectiveness of different imbalanced data management strategies within a machine learning framework for non-invasive SSC. The proposed approach relies exclusively on heart rate and motion signals, which can be acquired through wearable devices or contactless under-mattress sensors, making it suitable for longitudinal monitoring scenarios. Using the PhysioNet DREAMT dataset, 32 experimental scenarios are defined by combining data-level techniques (ADASYN oversampling with different balancing weights), algorithm-level strategies (cost-sensitive learning), and hybrid solutions. Four model families are evaluated—Decision Tree, k-Nearest Neighbors, Ensemble Classifiers, and Artificial Neural Networks—across classification tasks involving 2, 3, 4, and 5 sleep stages. The experimental results show that ensemble-based models provide robust and consistent performance under severe class imbalance, achieving macro accuracies of 82% for sleep–wake detection, 73% for 3-stage classification, 72% for 4-stage classification, and 64% for 5-stage classification. These findings confirm the relevance of imbalance-aware analytics and demonstrate the feasibility of accurate, minimally invasive SSC within big data and cognitive computing paradigms. Full article

The detection of sulfur hexafluoride (SF₆) decomposition products is critical for diagnosing insulation faults in gas-insulated switchgear (GIS). In this study, a vanadium-doping strategy was incorporated into the graphene/MoS₂ (GM) heterojunction to design a vanadium-doped graphene/MoS₂ (GMV) heterojunction material. Leveraging first-principles density functional theory (DFT), the adsorption behaviors of five characteristic SF₆ and its decomposition gases (H₂S, SO₂, SOF₂, SO₂F₂) on intrinsic GM and GMV were systematically investigated to evaluate their potential for gas sensing applications. Computational results reveal that intrinsic GM exhibits only weak physical adsorption toward all target molecules, with low adsorption energies and negligible charge transfer, which fails to meet practical application requirements. In contrast, GMV demonstrates significantly enhanced adsorption energies for H₂S, SO₂, and SOF₂ at vanadium sites (with a maximum value of −0.388 eV for SO₂) and shorter adsorption distances, while SO₂F₂ and SF₆ preferentially adsorb near electron-deficient carbon regions. Intrinsic GMV displays semimetallic properties, with a Fermi level at 0.126 eV and a band gap of 0.0017 eV. Upon adsorption of H₂S, SOF₂, SO₂F₂, or SF₆, the Fermi level undergoes a moderate shift (ranging from −1.083 eV to +0.349 eV), with minimal changes in the band gap. Conversely, SO₂ adsorption induces a substantial downward shift of the Fermi level to −1.732 eV, accompanied by the emergence of a sharp partial density of states (PDOS) peak near the Fermi level (0–1.5 eV), indicating strong orbital coupling and significant charge transfer. Furthermore, recovery times calculated using classical formulas show that at room temperature and a frequency of 1 × 10⁶ Hz, the recovery time of GMV for SO₂ is 2.43 s, outperforming the other four gases and satisfying practical gas sensing requirements. Through comprehensive analysis of adsorption distances, electronic structure changes, and recovery times, GMV exhibits higher selectivity toward SO₂. Thus, GMV can serve as a sensing material for detecting GIS insulation faults associated with elevated SO₂ concentrations, offering a viable strategy for advancing online monitoring technologies in power systems. Full article

Mercury ions (Hg²⁺) are highly toxic and pose severe risks to human health and ecosystems, necessitating sensitive detection methods for environmental monitoring. Here, we report a paper-based graphene sensor functionalized with single-stranded DNA (ssDNA) probes for Hg²⁺ detection based on T-Hg²⁺-T coordination chemistry. To elucidate the effect of probe structure on sensing performance, we designed DNA constructs with varying numbers of guanine (G) bases (3–6, designated DNA2–DNA5) in the bridging fragment and systematically evaluated their influence on hairpin stability, Hg²⁺ binding affinity, and sensor response. The DNA3-based sensor (four G bases) exhibited optimal electronic stability and sensitivity, achieving a detection limit of 0.673 pM with effective real-time monitoring capability in aqueous media. These findings highlight the critical role of DNA sequence design in T-Hg²⁺-T-based biosensors and provide a promising strategy for sensitive and selective Hg²⁺ detection in environmental samples. Full article

Monitoring thermal effluent is critical for assessing aquatic ecosystem health, yet traditional satellite remote sensing and in situ point measurements often fail to capture fine-scale thermal dynamics in narrow streams and complex coastal areas due to spatiotemporal resolution limitations. This study establishes a high-precision surface water temperature mapping protocol using a low-cost Unmanned Aerial Vehicle (UAV) equipped with an uncooled thermal infrared sensor (FLIR Vue Pro R) to overcome these observational gaps. We investigated two distinct hydrological environments—an inland stream and a coastal sea—to provide initial evidence for the applicability of an in situ-based linear regression calibration model across contrasting aquatic settings. The initial uncalibrated radiometric temperatures exhibited significant bias errors reaching up to 9.2 °C in the stream and 9.4 °C in the coastal area, primarily driven by atmospheric attenuation and environmental factors. However, the proposed calibration method dramatically reduced these discrepancies, achieving Root Mean Square Errors (RMSE) of 0.43 °C and 0.42 °C, respectively, with high determination coefficients (R² > 0.87). The derived high-resolution thermal maps successfully visualized the detailed diffusion patterns of thermal plumes, revealing a steep temperature gradient of approximately 13 °C in the stream discharge zone and a distinct 5 °C elevation in the coastal effluent area relative to the ambient water. These findings demonstrate that UAV-based thermal remote sensing, when coupled with a rigorous radiometric calibration strategy, can serve as a cost-effective and reliable tool for environmental monitoring, bridging the critical scale gap between local point measurements and regional satellite observations. Full article

Odometry estimation, which calculates the trajectory of a moving object across timeframes, is a critical and time-consuming function in SLAM (Simultaneous Localization and Mapping) systems. Although LiDAR-based sensing is most popular for outdoor and long-range applications because of its ranging accuracy, the sparsity of laser point cloud poses a significant challenge to feature extraction and matching in odometry estimation. In this paper, we investigate odometry estimation from two aspects, i.e., algorithm optimization, and system design/implementation. In algorithm optimization, we present an image feature-assisted odometry estimation scheme that leverages the richness of image information captured by a companion camera to enhance the accuracy of laser point cloud matching. This also serves as a screening mechanism to reduce the matching size and lower the computing complexity for a higher estimation rate. In addition, various schemes, such as adaptive threshold in image feature point selection, principal component analysis (PCA)-based plane fitting for laser point interpolation, and Gauss–Newton optimization for calculating the transform matrix, are also employed to improve the accuracy of odometry estimation. The performance of improved odometry estimation is verified using an existing FLOAM (Fast Lidar Odometry and Mapping) framework. The KITTI dataset for autonomous vehicles with ground truth was used as the test bench. Simulation results indicate that the translation error and rotation error can be reduced by 16.6% and 1.3%, respectively. Computing complexity, measured as the software execution time, also reduced by 63%. In system implementation, a hardware/software (HW/SW) co-design strategy was adopted, where complexity profiling was first conducted to determine the task partitioning and time-consuming tasks are offloaded to a hardware accelerator. This facilitates real-time execution on a resource-constrained embedded platform consisting of a microprocessor module (Raspberry Pi) and an attached FPGA board (Pynq Z2). Efficient hardware designs for customized DSP functions (adaptive threshold and PCA) were developed in an FPGA capable of completing one data frame in 20ms. The final system implementation met the target throughput of 10 estimations per second, and can be scaled up further. Full article

Electromagnetically induced transparency-like effects in silicon metasurfaces have attracted considerable interest due to their capability to manipulate optical resonances and improve sensing performance. In this work, a U-shaped silicon metasurface is proposed, consisting of a horizontal nanopillar supporting bright mode and two vertical nanopillars supporting dark mode. The coupling and coherent interference between the bright and dark modes lead to a pronounced EIT-like effect at specific wavelengths. By introducing nanoscale gaps between the horizontal and vertical silicon pillars, a U-shaped silicon metasurface with gap mode (UG metasurface) is formed, which induces strong near-field enhancement and is associated with reduced radiative losses, thereby improving the quality factor of the EIT-like resonance of UG metasurfaces. Two silicon metasurface samples are fabricated, and their transmission spectra are experimentally measured, showing good agreement with numerical simulations. In addition, the refractive index sensing performance of silicon metasurfaces is numerically investigated. The results show that the UG metasurface design significantly enhances the sensing capability, increasing the figure of merit from 6 RIU⁻¹ to 60 RIU⁻¹. The proposed silicon metasurfaces and near-field enhancement with the gap-mode mechanism provide a promising strategy for realizing high-performance optical sensing and offer valuable insights into the manipulation of electromagnetic responses. Full article

The precise determination of highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) energy levels is critical for understanding the photophysical and photochemical properties of carbon dots (C-dots), which directly govern their performance in optoelectronic, catalytic, and sensing applications. However, the lack of distinct redox peaks in cyclic voltammetry (CV) curves of C-dots poses a major challenge to conventional energy level calculation methods. Herein, we propose a novel strategy to calculate the HOMO–LUMO energy levels of C-dots by combining electron transfer (ET) kinetics with Marcus theory. A series of quinones (electron acceptors, EAs) and ferrocene derivatives (electron donors, EDs) were employed to quench the fluorescence of C-dots, and the ET rate constants (K) were derived from fluorescence lifetime measurements. The CV curves of EAs and EDs provided their respective oxidation and reduction potentials, which were used as reference energy levels. The UV–Vis absorption spectra confirmed that the fluorescence quenching mechanism was dominated by ET rather than energy transfer. Based on Marcus theory, the free energy change (ΔG) of ET reactions was correlated with K, and the HOMO and LUMO energy levels of C-dots were calculated to be −1.84 V (vs. SCE) and +1.60 V (vs. SCE), respectively. This study not only provides a reliable method for determining the energy levels of C-dots without distinct redox peaks but also deepens the understanding of ET mechanisms between C-dots and small molecules. The proposed strategy is expected to be extended to other fluorescent nanomaterials with similar CV limitations. Full article

The use of biostimulants in agriculture is increasing; however, their effects on raspberry remain insufficiently understood. The aim of this study was to evaluate the impact of foliar-applied biostimulants on yield and growth in three primocane raspberry cultivars grown under field conditions using multispectral imaging based on unmanned aerial vehicles. An experiment included a control and four foliar biostimulant treatments based on animal-derived amino acids, plant-derived amino acids, seaweed extract, and seaweed extract combined with animal-derived amino acids. Biostimulant effects on primocane raspberry were found to vary substantially depending on cultivar, environmental conditions, and formulation type, with measurable impacts on both yield formation and vegetative growth. These responses were further supported and characterized using multispectral UAV-based mutlispectral imaging, which enabled effective detection of treatment-related physiological changes. This approach was based on the analysis of relative percentage changes between consecutive measurements of selected vegetation indices, allowing the identification of dynamic physiological responses over time. These findings highlight the need for a more targeted approach to biostimulant use, taking into account cultivar-specific responses and environmental variability. Future research should extend this framework to a broader range of genotypes, cultivation systems, and biostimulant formulations, while integrating remote sensing with other analytical methods to better understand plant physiological responses. Such developments may support the transition toward data-driven and precision-guided biostimulant application strategies in sustainable crop production. Full article

Integrated sensing and communications (ISAC) requires tight coordination between spatial signal design and multiple-access strategies to balance communication throughput and sensing accuracy under shared spectral and hardware constraints. However, existing ISAC frameworks with rate-splitting multiple access (RSMA) typically rely on fixed antenna arrays and decoupled optimization, which fundamentally limit their ability to adapt to fast channel variations and dynamic sensing requirements. This paper introduces a fluid antenna-enabled RSMA-assisted ISAC architecture, in which movable antenna ports are exploited as a new spatial degree of freedom to enhance adaptability in both communication and sensing operations. Fluid antenna systems (FAS) are deployed at both the base station and user terminals, allowing dynamic port selection that reshapes the effective channel and sensing beampattern in real time. We formulate a joint sum-rate maximization problem subject to explicit sensing-quality constraints, capturing the coupled impact of antenna port selection, RSMA rate allocation, and multi-beam transmit design. The proposed framework maximizes the communication sum-rate while ensuring that the sensing functionality satisfies a predefined sensing quality constraint. This constraint-based ISAC formulation guarantees that sufficient sensing power is directed toward the target while optimizing communication performance. The resulting optimization involves strongly coupled discrete and continuous decision variables, rendering conventional optimization methods ineffective. To address this challenge, a hierarchical deep reinforcement learning (HDRL) framework is developed, where an upper-layer deep Q-network (DQN) determines discrete antenna port selection and a lower-layer twin delayed deep deterministic policy gradient (TD3) algorithm optimizes continuous beamforming and rate-splitting parameters. Numerical results demonstrate that the proposed approach significantly improves system performance, achieving higher communication sum-rate while satisfying sensing requirements under dynamic propagation conditions. Full article

Land cover (LC) change is reshaping terrestrial ecosystems and profoundly impacting sustainable development in Africa, yet the long-term, continental-scale spatiotemporal dynamics of these shifts remain obscured. To address the above issue, this study systematically explores the spatiotemporal dynamics of LC across Africa from 1985 to 2022 by leveraging the fine-resolution remote-sensing-derived GLC_FCS30D LC dataset within a stratified Intensity Analysis framework. To decompose landscape changes into interval, category, and transition levels across five climatic sub-regions of Africa, we systematically evaluate the temporal consistency of land systems. This hierarchical approach disentangles systematic transition pathways from random fluctuations, thereby revealing the distinct regional regimes governing continental transformation of LC. Our results ultimately show a strong LC change acceleration in Africa after 2010, mainly in Southern, Eastern, and Western Africa, which together made up 80 to 90% of the continent’s LC dynamics. During the whole study period, shrubland and grassland had the highest gross turnover due to their high bidirectional volatility. Intensity-wise, forest remained inactive even though it was a persistent net loser to crop in East Africa (2010–2020), to shrub in Southern Africa (1990–2022), and to wetland in West Africa during the post-2000 intervals. Wetland had a major change in dynamics from historical growth during 1985–1990 to systematic decline in 2015–2022. Cropland increased by systematically targeting shrubland and grassland, mainly in East Africa. Additionally, the Sahel contributed 40% of continental grassland to bare area transitions, despite some recovery of grassland in the region. These findings show that aggregate net-change metrics obscure the volatility in African LC; therefore, distinct regional regimes such as agricultural expansion and forest degradation necessitate spatially differentiated management strategies. Full article

Reliable underwater acoustic (UWA) communication is fundamental to marine sensing applications, including environmental monitoring, underwater sensor networks, and autonomous platforms, yet remains severely challenged by multipath propagation, Doppler effects, and limited bandwidth. This paper investigates a memory-based multicarrier modulation framework in which controlled phase continuity is introduced at the symbol-mapping stage to enhance robustness against channel-induced distortions. Unlike conventional memoryless multicarrier schemes, the proposed approach embeds intentional phase memory at the transmitter and exploits it at the receiver, improving reliability in highly dispersive underwater environments. A comprehensive bit-error-rate (BER) evaluation is conducted using extensive simulations over realistic shallow-water acoustic channel models. The analysis examines rational modulation indices, pulse-shaping filters, roll-off factors, transmitter–receiver separation distances, and receiver structures. Both matched-filter and zero-forcing receivers are considered to assess trade-offs between interference mitigation and noise amplification. Results demonstrate consistent and significant BER improvements compared with conventional memoryless multicarrier systems. A modulation index of 7/16 achieves the minimum BER with matched-filter detection, while 3/10 yields optimal performance with zero-forcing detection. The Dirichlet pulse provides the most robust performance across operating conditions. These findings establish phase-memory-aware multicarrier design as a practical strategy for reliable underwater sensing and communication systems. Full article