Search Results (844)

Feature-based visual–inertial odometry (VIO) often suffers from initialization failures and tracking drift under degraded visual conditions, such as low-texture regions, abrupt illumination changes, and scenes with a high ratio of dynamic correspondences. We present RISE-VIO, a real-time inertial-navigation-system-centric (INS-centric) visual–inertial odometry system that improves robustness by introducing GNC-style robustification into two failure-critical stages: initialization and per-frame pose estimation. For robust initialization, we develop a GNC-based decoupled rotation–translation initialization module with a two-stage observability gate, consisting of (i) rotation-compensated parallax-rate screening and (ii) a spectral-stability test on the linear global translation (LiGT) system. For online robustness, we design an IMU-prior-guided GNC-EPnP module to selectively downweight or reject outlier correspondences during pose estimation. Experiments on public benchmark datasets show that RISE-VIO achieves more reliable initialization and more stable trajectory estimation in challenging visual conditions while maintaining real-time performance. Additional Monte Carlo perspective-n-point (PnP) evaluations further support the robustness of the proposed pose estimation module under severe outlier contamination. Full article

Train track occupancy detection is essential for railway operation safety and dispatching, yet GNSS-based positioning and track matching can degrade or fail in turnouts and station yards due to multipath, interference, and dense track layouts. This paper presents an IMU-only method to discriminate track-switching events during turnout passage by exploiting the transient change in heading angular velocity. The Z-axis gyroscope measurement (approximately aligned with the track-plane normal) is used as a heading-rate proxy, and a lightweight indicator is constructed from the difference between a short-window moving average and the full-run mean. The full-run mean further serves as an in situ approximation of the gyroscope zero bias, alleviating the need for pre-calibration and improving robustness to systematic drift. A fixed discrimination threshold is determined from stationary gyroscope noise statistics, and the minimum effective operating speed is derived by combining gyro noise characteristics with the kinematic relationship among train speed, turnout curvature radius, and heading rate. Field experiments conducted from January to April 2025 on three railway sections covering 27 turnouts (300 turnout-passage events) show that, using a constant threshold $T_0=0.002\mathrm{rad}/\mathrm{s}$, the proposed method achieves 100% track-switching discrimination accuracy within 5–40 km/h, without requiring track maps, GNSS, or prior databases. Full article

To investigate the movement characteristics of fish eggs and to clarify their movement behavior as flow conditions transition from slow to rapid, a hydrodynamics-based fish egg movement model was proposed. Indoor flume experiments under slow-flow conditions conducted previously by the research group were used as a basis. Twenty-three operating conditions with different water depths and discharges were designed, including fifteen rapid-flow conditions and eight slow-flow conditions. Numerical simulations were performed to examine the influence of flow velocity on fish egg movement under different flow conditions. The results show that fish eggs drift with the flow under different flow conditions, and their longitudinal velocity lags behind the flow velocity. At low flow velocities, the vertical velocity distribution of fish eggs is relatively concentrated. With increasing flow velocity, the vertical velocity becomes more dispersed in the high-velocity range. A power–law relationship exists between flow velocity and the trajectory slope of fish egg movement. When the flow velocity is lower than 0.5 m/s, the trajectory slope varies significantly with flow velocity; when it exceeds 1.2 m/s, the slope approaches a constant value. Water depth has a limited influence on fish egg velocity and trajectory slope under both slow-flow and rapid-flow conditions. By combining the relationships among flow velocity, trajectory slope, and suspension rate, a flow velocity of 0.3 m/s is identified as the critical flow velocity for maintaining the safe drifting of fish eggs. The findings provide technical support for ecological operation strategies aimed at fishery resource conservation. Full article

The navigation conditions of inland river crossing waterways are directly related to the efficiency and safety of the entire water transport network. In this paper, a two-dimensional hydrodynamic model is established by using Delft3D to simulate the crossflow distribution characteristics before and after the excavation project under the condition of 98% guaranteed flow rate (1690 m³/s). On this basis, the optimized channel width calculation formula is introduced to quantify the drift of ships of different tonnage classes (1000 t and 2000 t) under the action of crossflow. The results show that the maximum lateral flow velocities of north branch, middle Branch and south branch after excavation are 0.57 m/s, 0.42 m/s and 0.50 m/s. Based on the calculation results of the required channel width and the actual situation of the section, the organizational scheme of adopting one-way navigation under the condition of high flow during the flood season is proposed, and the speed of downbound ships (1000 and 2000 t) should not be less than 9 km/h to ensure the safety of one-way navigation. In the upbound ship, the 1000-t class needs to be not less than 6 km/h, and the 2000-t class needs to be not less than 7 km/h. The study establishes an engineering-oriented quantitative link from hydrodynamic cross-current analysis to navigation-width assessment and further to traffic organization under flood-season conditions, providing practical support for navigation safety management in complex inland river confluence reaches. Full article

Sea-surface reflections and wind–wave motion render maritime channels strongly time-varying and statistically non-stationary, while nearshore deployments face sparse infrastructure and co-channel multiuser interference. This study integrates reconfigurable intelligent surfaces (RISs) with rate-splitting multiple access (RSMA) for joint online resource allocation. A physics-inspired time-varying channel model is established by embedding fractional Brownian motion-driven slow statistical drift and reflection-phase perturbations. With imperfect, delayed channel state information (CSI) and discrete RIS phase quantization, a proportional-fairness utility maximization problem is formulated to jointly optimize shore base-station precoding, RIS phase shifts, and RSMA common-rate allocation. To cope with strong non-convexity, high dimensionality, mixed continuous–discrete coupling, and partial observability, a fractal-aware recurrent robust Actor–Critic (FRRAC) algorithm is developed. FRRAC encodes short observation histories using a gated recurrent unit and incorporates a lightweight Hurst-proxy estimator to capture slow channel statistics for robust value evaluation and policy learning. Truncated quantile critics and mixed prioritized–uniform replay further improve value robustness, training stability, and sample efficiency. Simulation results show that FRRAC converges faster and more stably under both conventional and fractal non-stationary channel modeling, and outperforms representative baselines across the objective and multiple statistical metrics, validating its effectiveness for joint resource optimization in maritime RIS–RSMA systems. Full article

Existing autonomous navigation systems for Unmanned Aerial Vehicles (UAVs) face the dual challenges of local minima entrapment and computational complexity that scales with environmental density. This paper proposes a hierarchical navigation architecture integrating deep representation learning with an improved Vortex Artificial Potential Field (APF). At the decision layer, a Convolutional Neural Network (CNN) encodes the environment as a fixed-dimensional tensor and generates global waypoints with constant-time inference, independent of obstacle count. At the control layer, a Vortex APF resolves the Goal Non-Reachable with Obstacles Nearby (GNRON) problem and limit-cycle oscillations through tangential rotational potentials, achieving significant improvement in trajectory smoothness compared to traditional APF methods. A closed-loop replanning mechanism further ensures robust performance under execution drift. Experiments across varying obstacle densities demonstrate that the combined system achieves high navigation success rates in dense environments with substantially reduced computation time compared to sampling-based planners such as Rapidly exploring Random Tree star (RRT*), while maintaining superior trajectory quality. This architecture provides a computationally efficient solution for resource-constrained UAV platforms operating in GPS-denied or obstacle-rich environments such as warehouses, forests, and disaster sites. Full article

This paper investigates the strong convergence of jump-diffusion processes with Markovian switching using the truncated Euler–Maruyama (TEM) method. Under the assumption that the drift and diffusion coefficients satisfy a Khasminskii-type condition and the jump coefficient meets a linear growth condition, we derive the convergence rate. Furthermore, we demonstrate that the TEM method effectively preserves both the mean square stability and the asymptotic boundedness of the underlying jump-diffusion process. A case study involving music signals is provided to illustrate the theoretical findings. Full article

We propose Distribution-Preserving Latent Steganography via Conditional Optimal Transport (DPL-COT), a coverless image steganography framework for latent diffusion models. Unlike classical cover-modifying schemes, DPL-COT embeds a bitstream directly into the initialization noise latent ${\mathbf{z}}_T\sim \mathcal{N}(\mathbf{0},\mathbf{I})$ without model retraining. Our primary objective is high recoverability and a low bit error rate (BER) under deterministic inversion, which is inherently imperfect due to numerical discretization and VAE nonlinearity. To maximize decoding stability, we restrict embedding to the natural tails of the latent prior by selecting the largest-magnitude coordinates, thereby increasing the sign decision margin against inversion drift. To preserve distributional stealth, per-bit target values are analytically derived from truncated Gaussians matching the marginal distribution of the selected coordinates. Conditional 1D optimal transport is applied independently for each bit class, mapping every coordinate to its target value while preserving rank order. We generate 5000 stego images using a pretrained diffusion model and demonstrate a favorable capacity–reliability trade-off (e.g., 4916 bits/image with 0.473% mean BER) and strong robustness to JPEG compression (sub-1% mean BER at $Q=60$). Compared with LDStega, a recent LDM-based scheme reporting 99.28% clean-channel accuracy, DPL-COT achieves 99.53% at a comparable operating point and sustains above-99% accuracy under all tested JPEG quality factors. Latent-space tests further confirm negligible cover–stego distribution shift (mean ${\mathrm{KS}}_2<0.003$, mean $W_1<0.003$), a property not formally addressed by prior methods. Full article

Respiration rate (RR) is a key physiological indicator of health, stress, and thermoregulatory load in dairy cattle, yet continuous RR monitoring under commercial farm conditions remains challenging. In this Technical Note, we present a non-invasive clip-on nose ring device for continuous respiration monitoring based on acoustic recording directly at the nostril. The device integrates a MEMS microphone, embedded electronics, battery, and removable storage in a sealed, mechanically robust housing suitable for real-world barn environments. The system was deployed on five dairy cows under commercial farm conditions, enabling repeated multi-day recordings over several weeks. The respiration rate was extracted offline from raw audio using a deterministic signal-processing pipeline based on multiscale periodicity detection. Algorithm-derived RR estimates were evaluated against manually annotated breath events. Using 10-min rolling median values, the algorithm achieved a mean absolute error (MAE) of 1.47 breaths per minute (bpm), a root mean square error (RMSE) of 1.92 bpm, and a high correlation with reference values (r = 0.98, R² = 0.96). In addition to short-term accuracy, the system enabled stable multi-day monitoring. Group-level analysis across all five animals revealed a clear diurnal respiration pattern over multiple consecutive days, with lower RR during nighttime and higher RR during daytime summer conditions, without signs of a baseline drift. These results demonstrate the feasibility of continuous, long-term respiration monitoring in dairy cattle using an audio-based clip-on nose ring device and provide a practical foundation for longitudinal (multi-day, within-animal) RR assessment under commercial farm conditions, with potential for future extensions towards advanced respiratory health monitoring. While the system demonstrated stable performance under summer farm conditions, validation under extreme heat-stress environments and larger animal cohorts is required for comprehensive population-level assessment. Full article

To achieve high-precision and real-time quantitative evaluation of gas production in underground gas storage (UGS), this study focused on 11 typical injection-production wells in the Sudong UGS group. To address the common challenges posed by deviated well structures and complex wellbore temperature field distributions, a gas flow-rate calculation method based on Distributed Temperature-Sensing (DTS) data was developed. By standardizing the processing of multi-well temperature data, deviated wellbore trajectories were straightened to convert measured depth (MD) to true vertical depth (TVD). By incorporating a geothermal correction mechanism, temperature anomalies closely related to fluid flow were extracted, and a spatially unified temperature field model was constructed. On this basis, a “Dual-Point Temperature Difference Method” is proposed as a novel approach for single-well production evaluation. Based on thermodynamic phenomena such as the Joule–Thomson effect and expansion cooling, two critical sensing points, upstream and downstream of the production layer, were selected, with their temperature anomaly difference (∆T) serving as a sensitive indicator of flow rate variations. Combined with downhole pressure parameters and synchronized wellhead metering data, a nonlinear quantitative relationship model between ∆T and gas production rate Q was established, enabling accurate conversion of wellbore thermal response to macroscopic flow parameters. The results indicated that the gas production rates calculated by this method align well with traditional wellhead metering data, with errors maintained within engineering tolerances. Notably, the method demonstrates higher reliability and corrective capabilities in wells with drifting or faulty meters. This achievement breaks the reliance of traditional methods on specific layers or mechanical meters. It enables the effective application of multi-well, full-section, and non-contact temperature data in gas volume assessment. This research provides new technical support for dynamic monitoring, efficient operation, and remaining gas evaluation of UGS, offering significant prospects for engineering applications. Full article

Patients receiving dialysis treatments suffer from a high rate of systemic comorbid conditions, including cardiovascular disease, mineral and bone disorders, chronic inflammation, amyloidosis, and recurring infections, leading to increased morbidity and mortality rates despite the progress made in the field of renal replacement therapies. The aforementioned conditions result from the continued dysregulation and overproduction of molecular biomarkers, which cannot be adequately monitored by traditional, intermittent laboratory tests. This review critically assesses the newly developed biosensor technologies for the detection of major dialysis biomarkers, including potassium, phosphorus, parathyroid hormone (PTH), β₂-microglobulin, creatinine, and cystatin C, with special emphasis on biosensors based on electrochemistry, optics, impedimetry, nanophotonics, and biological engineering techniques. These recent biosensors have been evaluated based on their analytical performance, the biofluids used in the studies, and their suitability for measuring relevant concentrations of these biomarkers. Special attention is given to biosensors capable of continuous operation or minimally invasive sampling, as well as to newly developed biofluid sampling techniques, including microneedle-, microtube-, and micropillar-based systems, for the long-term monitoring of the biomarkers in the serum of patients receiving dialysis treatments. The biosensing techniques for measuring infection biomarkers have also been discussed, given the high risk of bloodstream and access infections among patients receiving dialysis. The limitations of these biosensors include biofouling, calibration drift, and their integration into the dialysis treatment workflow. Finally, the future prospects of the recent biosensors offer the possibility of the proactive management of the high rate of comorbid conditions in this high-risk population of patients receiving dialysis treatments. Full article

Continuous healthcare monitoring systems generate non-stationary physiological data streams, where evolving statistical properties and patterns often invalidate static models and fixed user classifications. To address this challenge, we propose drift-aware adaptive architecture that integrates concept drift detection into a distributed edge–cloud data analytics pipeline. In the proposed design, a concept drift is elevated from a maintenance signal to the primary mechanism governing user-state adaptation, model evolution, and inference consistency. Within the proposed system, the edge tier performs low-latency inference and preliminary drift screening under strict resource constraints, while the cloud tier executes advanced drift detection and validation, orchestrates user reclassification and model retraining, and manages model evolution. A feedback loop synchronizes edge and cloud operations, ensuring that detected drift triggers appropriate system transitions, either reassigning a user to an updated state category or initiating targeted model updates. This architecture reduces reliance on static group assignments, improves personalization, and preserves model fidelity under evolving physiological conditions. We analyze the drift types most relevant to healthcare data streams, evaluate the suitability of lightweight and cloud-grade drift detectors, and define the system requirements for stability, responsiveness, and clinical safety. Evaluation across 21 concurrent users demonstrates that drift-aware adaptation reduced prediction MAE by 40.6% relative to periodic retraining, with an end-to-end adaptation latency of 66 ± 37 s. Hierarchical cloud validation reduced the false-positive retraining rate from 88.9% (edge-only triggering) to 27.3%, while maintaining uninterrupted inference throughout all adaptation events. Full article

Background: Acute hepatitis E (AHE) poses escalating risks to older adults (≥65 years), compounded by immunosenescence and comorbidities. Using Global Burden of Disease (GBD) 2021 data, this study analyzes global AHE burden, trends, and projections in aging populations. Methods: Age-standardized rates (ASIR, ASMR, ASDR) for AHE in adults ≥ 65 years were extracted from GBD 2021 across 204 countries (1990–2021). Frontier analysis assessed gaps between observed burdens and sociodemographic index (SDI)-based theoretical minima. Age-period-cohort (APC) modeling evaluated age/period/cohort effects. Bayesian (BAPC), NORDPRED, and ARIMA models projected trends to 2050. Results: Global ASIR increased by 1.5% annually (1990–2021), with ASMR and DALYs declining significantly. Middle SDI regions showed the steepest ASIR rise (net drift: 0.064%/year), while high SDI areas had volatile trends. Age effects peaked in ≥95-year-olds. Frontier analysis revealed persistent ASIR-SDI gaps, particularly in low-middle SDI regions. Projections indicate a ASIR rise by 2050 (113.04/100,000), contrasting with declining ASMR (0.056/100,000) and ASDR (1.31/100,000) and the NORDPRED, ARIMA, and EAPC models exhibit analogous global predictive trends. Conclusions: Diverging trends of rising incidence and falling mortality highlight unmet prevention needs. High-burden regions require SDI-stratified strategies, prioritizing vaccination programs (e.g., HEV 239), zoonotic transmission control, and enhanced surveillance. The Sustainable Development Goals (SDGs) envision hepatitis elimination by 2030 (Target 3.3). However, our analysis projects ongoing AHE burden in aging populations through 2050, indicating the need for post-2030 policy adaptations. Full article

Zero-trust architecture (ZTA) requires continuous and adaptive identity authentication to maintain security in dynamic environments. However, current federated learning (FL)-based authentication models often struggle to incorporate evolving attack patterns without experiencing catastrophic forgetting. Moreover, non-independent and identically distributed (non-IID) client data and concept drift frequently lead to degraded model robustness and personalization. To address these issues, this paper presents a hybrid learning framework that integrates federated learning with incremental learning (IL) for sustainable authentication. A Dynamic Weighted Federated Aggregation (DWFA) algorithm is developed to mitigate concept drift by adjusting aggregation weights in real time, ensuring that the global model adapts to changing data distributions. This approach enables continuous learning from distributed threat data while maintaining privacy and eliminating the need for historical data retention. Experimental results on real-world traffic datasets indicate that the proposed framework outperforms conventional FL baselines, reducing the overall error rate by approximately 56% and improving the detection rate for novel attack types by over 17.8%. Furthermore, the framework remains stable against performance decay while maintaining efficient communication overhead. This study provides an adaptive, privacy-preserving solution for identity authentication in zero-trust systems. Full article

Backround and Objectives: Basketball performance is shaped by repeated high-intensity actions interspersed with brief recovery. Conventional continuous or strictly incremental testing may not fully capture short active-recovery dynamics relevant to stop-and-go sports. Material and Methods: This study applied a VT₂-referenced progressive–intermittent treadmill protocol and focused on 60-s active-recovery kinetics to describe effort tolerance in an applied basketball setting. Basketball players from Mureș County completed anthropometry (24 h pre-test, fasted) and a single laboratory visit. Pre-test training and diet were standardized for 48 h (submaximal training; predominantly carbohydrate intake). CPET was performed in 3-min stages (6.5 km·h⁻¹ start; +0.7 km·h⁻¹ per stage) and stopped at RER = 1.00 and/or blood lactate = 4.0 mmol·L⁻¹ (operational VT2). After 3 min active recovery, participants completed six 60-s high-speed bouts separated by 60-s active recovery intervals (AR1–AR6), with intensities prescribed at 120–180% of VT2-derived speed, followed by an 8-min active recovery. For each AR interval, linear regression over 0–60 s yielded slopes for VO₂, VO₂/HR, VCO₂, V̇E, VE/VO₂, VE/VCO₂, and PetCO₂. Results: VT₁ was determined at 2.29 m·s⁻¹ (VO₂ 32 mL·min⁻¹·kg⁻¹) and VT2 at 3.07 m·s⁻¹ (VO₂ 42 mL·min⁻¹·kg⁻¹). Maximal intermittent speed was 5.33 m·s⁻¹ (VO₂ 45.5 mL·min⁻¹·kg⁻¹; RER 1.06; PetCO₂ 38 mmHg). VO₂ differed across successive bouts (p = 0.0001), while PetCO₂ showed a small downward drift across repetitions. Peak indices (max speed, VE/VCO₂max, PetCO₂max, VEmax) were associated with phase-specific recovery slopes across early, mid, and late recovery periods (false discovery rate–adjusted correlations). Lactate decreased over 8 min, but lactate change rates were not associated with peak indices. Conclusions: The VT₂-referenced progressive–intermittent protocol appears feasible in basketball players and provides phase-dependent recovery information that complements conventional peak CPET outcomes, with potential relevance for applied team settings. Full article