Search Results (909)

Nonlinear filtering can be formulated as the propagation and update of conditional probability densities, but direct numerical propagation of the associated Forward Kolmogorov equation (FKE) over a large fixed domain is computationally expensive. This paper proposes a Recentered-Domain Yau–Yau Filter (RD-YYF) with reduced-order FKE propagation for nonlinear state estimation. The method solves the FKE on a fixed-size local computational window centered at the latest state estimate, thereby concentrating numerical resolution near the dominant posterior density. In the offline stage, physics-informed neural networks (PINNs) generate FKE solution snapshots, principal component analysis constructs a low-dimensional representation of density evolution, and a lightweight residual surrogate maps initial-condition coefficients and the domain center to terminal-solution coefficients. In the online stage, the pretrained surrogate performs per-timestep density prediction within the recentered window, followed by observation update and state estimation. Numerical experiments on two geometrically constrained target-tracking models show that RD-YYF achieves lower tracking errors than the extended Kalman filter and particle filter under matched online evaluation conditions. A fixed-domain ablation further shows that recentering improves density approximation in high-probability regions and reduces offline PINN training epochs. These results indicate that recentered-domain reduced-order FKE propagation is a practical computational strategy for nonlinear density-based filtering. Full article

Reliable probabilistic prediction of dam-break wave run-up is important for screening downstream barriers and river infrastructures. However, it remains unclear whether propagation conditions exhibit stable statistical groups within the observed predictor domain and whether explicitly incorporating these groups improves prediction beyond the underlying continuous variables. This study develops a probabilistic framework integrating physics-guided self-tuning spectral clustering (PG-STSC), quantile regression forests (QRFs), and conformal calibration diagnostics. The framework was applied to 155 physical model observations characterized by relative propagation distance, Froude number, relative wave height, relative wavelength, and wave nonlinearity, with the relative maximum run-up as the response variable. Full-data PG-STSC identified two groups containing 66 and 89 observations, separating mainly shorter-distance, higher-$Fr$ conditions from longer-distance, lower-$Fr$ conditions. The partition showed moderate physical-space separation but high subsample stability, while the group-wise run-up distributions overlapped substantially. In fivefold out-of-fold evaluation, the group-free QRF achieved a continuous ranked probability score (CRPS) of 0.1537, corresponding to 22.8% CRPS skill relative to the unconditional empirical distribution. Global and group-conditioned conformal corrections produced identical pooled 90% coverage (0.942) and interval widths, with group-specific coverages of 0.922 and 0.952, respectively. Supported-domain exceedance maps translated the predicted conditional distributions into threshold screening information while masking unsupported extrapolation. Overall, physically coherent clustering improves interpretation and conditional diagnostics but does not necessarily enhance predictive performance. The resulting exceedance estimates are suitable for within-domain scenario screening rather than site-specific overtopping assessment. Full article

Grid-impact studies treat battery electric vehicles as loads, and ask when network capacity will be exhausted. This paper reverses the question: how much of the fleet must operate bidirectionally, and with what probability will an achievable participation rate suffice, for the network to remain within its limits? A conceptual framework adds a vehicle-to-grid and vehicle-to-building flexibility term to the balance between available and required power, nests the authors’ earlier deterministic model for twenty municipalities in Northern Portugal as its zero-flexibility special case, derives a closed-form break-even participation rate per municipality and year, and keeps the simultaneity assumption of that model explicit as a coincidence factor. Participation, location, plug-in and export parameters follow beta-PERT distributions calibrated on published trials and surveys, propagated by Monte Carlo simulation without new field data. The framework is an apparent-power balance per municipality, so its outputs are an upper bound on usable flexibility, not a feeder-level feasibility check. An enrolled vehicle provides about 11 kVA of peak relief, over nine tenths from not charging rather than exporting. Under worst-case simultaneity, observed participation rates, if in place from the outset, halve the 2028 shortfall probability but cannot prevent shortfall by 2030; under realistic coincidence the regional network is not constrained and only eight of twenty municipalities remain critical. The network balance is replicable wherever municipal substation data exist; behavioural parameters require local calibration. Full article

Published assessments of Bitcoin’s exposure to a cryptographically relevant quantum computer convert resource estimates into risk figures by substituting a point estimate of the key derivation time into an exponential tail. We show that this procedure is systematically optimistic. Because the exponential tail is strictly convex, its expectation over any non-degenerate break time distribution exceeds its value at the mean, so every such figure is a provable lower bound on the true risk: at an unchanged nine-minute mean, exponential dispersion moves Bitcoin’s on-spend theft probability from 41% to 53%. We develop the distributional model this requires, a race between a Poisson block-arrival process and a random time-to-key embedded in a Nakamoto reorganization contest and a replace-by-fee contest, and obtain closed forms for the theft probability, for the commit–reveal delay attaining a given security target, and for the coin value an owner retains in the fee war. Replacing the zero-delay catch-up bound with a delay-aware one raises the required delay by a factor of 1.2 to 20.4, a correction driven almost entirely by the adversary’s pre-mining lead rather than by propagation delay. Reconciling our results with a concurrent round-based analysis shows that an apparent threefold disagreement in the literature is a difference in security target, not in substance. Finally, we test the block-arrival assumption against 40,320 block headers: the exponential marginal law holds, but a conditional uniformity test detects within-epoch rate drift invisible to a Kolmogorov–Smirnov test, an effect worth under a third of a percentage point and again conservative. Full article

Non-cooperative sensing using signals of opportunity traditionally requires an explicit reference signal for target detection and localization. This paper introduces a reference-free sensing framework in which target geometry is inferred directly from the received waveform rather than by comparison with an acquired or reconstructed illuminator signal. The proposed framework is implemented using the Ranging, Detection, Imaging, Communications, Approach, and Landing (RaDICAL) architecture, which combines a hybrid Dish–Sparse Uniform Circular Array (SUCA) receiver with Starlink downlink transmissions as spaceborne illuminators of opportunity. Deterministic Multifrequency Dither (DMD) applied across the SUCA elements transforms spatial diversity into unique composite waveform signatures. A unified electromagnetic and signal-processing model is developed that combines spherical-wave propagation, parabolic focusing, deterministic multifrequency modulation, and QR-based waveform-domain hypothesis testing for direct target localization. Numerical simulations together with link-budget analysis demonstrate the feasibility of the proposed approach. Single-dwell detection of 0 dBsm targets is achieved at physical signal-to-noise ratios near 0 dB, while near-unity detection probability is obtained above 10 dB SNR under controlled false-alarm conditions. The results demonstrate that commercial Starlink LEO communication satellites can serve as practical illuminators of opportunity for reference-free non-cooperative sensing without requiring acquisition or reconstruction of the transmitted illuminator waveform. Full article

In comparison with stratovolcanoes, shield volcanoes tend to have more frequent dike-fed eruptions and large lateral and vertical collapses, as well as more gently dipping flanks. In many stratovolcanoes, dike-fed eruptions occur once every several hundred or thousand years but once every few years in many shield volcanoes. Using Hamilton’s principle of least action as a basis for determining potential dike/sheet propagation paths, it is shown that the probability of arrest of an injected dike/sheet is normally much greater in a stratovolcano than in a shield volcano. This is primarily because in stratovolcanoes rock layers and units are of contrasting mechanical properties, so that many dikes become arrested and thus do not feed eruptions. Similarly, many faults in stratovolcanoes become confined to one or several layers/units and do not reach the surface to generate landslides or ring faults. Consequently, the formation of large landslides is generally more difficult—requires more energy—in composite volcanoes than in shield volcanoes. For the same reason, formation of calderas in stratovolcanoes is normally more difficult than in shield volcanoes. More energy is needed to propagate fractures through many layers/units in stratovolcanoes than in shield volcanoes. It follows that stratovolcanoes tend to be tougher, more resistant to tectonic fracture propagation, and thus mechanically stronger than shield volcanoes. This may partly explain differences in the frequencies of dike-fed eruptions, large landslides, and caldera collapses between shield volcanoes and stratovolcanoes. Full article

To address the challenges of cumulative pseudo-label noise and substantial phenological differences across regions in cropping structure mapping, this study develops a phenological knowledge-guided weakly supervised semantic segmentation framework, termed PhenoStruct-WSF. The framework first constructs typical phenological curves for major annual cropping patterns and uses TWDTW to measure the phenological similarity between samples and templates. A two-class margin indicator, ∆_i, is then proposed, and phenological confidence is derived through probability calibration, after which high-confidence samples are selected as anchors under posterior-FDR control. Subsequently, the anchors are embedded into a structure-aware weakly supervised segmentation network, PhenoStruct-Net, forming a two-stage training strategy of “fidelity first, expansion later”, in which robust pseudo-label expansion is achieved through phenological confidence gating and adaptive weighting. Experiments were conducted in three representative study areas, namely Handan in Hebei Province, Jingzhou in Hubei Province, and Siping in Jilin Province. The anchor samples selected using posterior-FDR achieved accuracies above 94% in all three study areas. PhenoStruct-WSF consistently achieved overall accuracy (OA), mean intersection over union (mIoU), and macro F1-score (mF1) above 89%, 81%, and 89%, respectively, across the three regions, with an average OA improvement of 4.52 percentage points over the comparison methods. Sample sensitivity experiments indicate that the proposed method maintains a stable trend of accuracy improvement under different training sample sizes, demonstrating good adaptability to limited samples. The ablation experiments further show that phenological confidence anchors, posterior-FDR control, and the confidence-weighting mechanism can effectively suppress pseudo-label noise propagation and improve the robustness of crop mapping in complex agricultural landscapes. Full article

Reliable localization and mapping are critical for intelligent robotic systems operating in dynamic indoor environments, where pedestrians and other moving objects can lead to erroneous feature associations, map contamination, and accumulated trajectory drift. To address these challenges, this study proposes the Boundary-Protected Semantic-Geometric Dynamic-Probability (Boundary-SGDP) framework, an enhanced red–green–blue-depth (RGB-D) visual simultaneous localization and mapping (SLAM) system based on boundary-protected semantic–geometric dynamic-probability estimation. The proposed method combines instance-level semantic priors generated by the YOLO26n-seg detector, a segmentation-oriented model in the You Only Look Once (YOLO) family, and the Segment Anything Model 2 (SAM2) with morphological region decomposition and RGB-D depth-edge detection. Potentially dynamic regions are further divided into dynamic interiors, semantic boundary protection bands, and geometrically informative depth-edge regions. Semantic and geometric cues are integrated to estimate a dynamic score for each feature, which is subsequently propagated to the MapPoint level as a dynamic probability. During pose optimization, these probabilities are used to adaptively adjust the weights of reprojection constraints, thereby reducing the influence of motion-contaminated observations while preserving geometrically valuable features around object boundaries and occlusion regions. Unlike conventional hard semantic masking strategies, Boundary-SGDP provides a soft and adaptive mechanism for handling dynamic observations. Experiments conducted on four dynamic walking sequences from the TUM RGB-D benchmark demonstrate that the proposed method achieves lower absolute and relative trajectory errors than the original ORB-SLAM3 system, while retaining substantially more boundary-related features. The results confirm the effectiveness of semantic–geometric fusion and boundary protection for robust visual localization and mapping in dynamic indoor scenes, and demonstrate the potential of the proposed framework for practical autonomous navigation and intelligent perception applications. Full article

Surface-water networks evolve as depressions fill, spill, and connect, whereas conventional digital elevation model (DEM) routing often treats flats and depressions through preprocessing or complete filling. This study introduces selective depression filling (SDF), a terrain-based framework that generates routing states characterized by depression extent, network connectivity, and geometric fill-to-spill volume. SDF aggregates 8-connected DEM cells with identical elevations into irregular cells (ICs), preserving the lateral geometry of flats, lakes, and channels, and uses a catchment-to-destination area (C/D) ratio to control depression-filling levels. IC-D8 (IC-based single-flow-direction routing scheme) and IC-MFD (IC-based multiple-flow-direction routing scheme) routing are evaluated across four contrasting landscapes. Both methods avoid ambiguous cellwise routing across flat and low-gradient features. Depression filling changes connectivity and contributing-area propagation as depressions transition from internal termini to widened flow-path components at spillover. Filling depths define DEM-derived topographic accommodation space rather than available water-storage capacity. Differences in C/D ratios define a Contributing-Area Sensitivity Metric (CSM), a terrain-based routing contrast rather than flood probability. Because the C/D ratio and filling states are derived from terrain data without independent hydrological forcing, the uncoupled SDF is a terrain-morphological analysis framework, not a hydrological model; hydrologic interpretation and regional transferability require independent calibration, validation, and testing across regions. Full article

Oblique and Fork-shaped traces in ionosonde’s ionograms have been identified as manifestations of nighttime and daytime Medium-Scale Traveling Ionospheric Disturbances (MSTIDs), respectively. The MSTIDs associated with these singular patterns have been verified by analyzing the evolution of detrended total electron content (dTEC) perturbation maps derived from the Global Navigation Satellite System (GNSS) signals. This has made it possible to create a database of MSTIDs characterized by their dominant period, horizontal propagation speed, wavelength, and propagation direction. This also allowed comprehensive case studies and statistical analyses of their temporal occurrence, direction of propagation, seasonal dependence, and its association with Spread-F (SF) phenomena. The results reveal that approximately 92% of the “Oblique” traces and 86% of the “Fork” traces in ionograms correspond to nighttime and daytime MSTID activity, respectively, as observed in the dTEC maps. Oblique traces are predominantly observed during southwestward propagating MSTIDs and exhibit a marked seasonal occurrence during summer and winter. Fork traces occurred predominantly during equatorward propagating MSTIDs and are most frequently observed in winter. A statistically significant association has been found between daytime MSTID activity (Fork traces) and the subsequent appearance of SF, with conditional probabilities ranging from ~87% to ~78% at mid-to-high latitudes and from ~76% to ~61% at mid-to-low latitudes. However, when overall MSTID activity was considered—taking into account both the disturbances associated with the daytime Forks and those associated with nighttime Oblique traces—the statistical association with subsequent occurrence of SF increased to approximately 91%. These results show a strong statistical relationship but do not establish a direct causal link between an individual daytime MSTID activity and nighttime SF. Instead, the occurrence of MSTID is seen as an indicator of ionospheric conditions that are favorable for the development of SF at mid-latitudes. These findings offer new insights into the ionogram-based characterization of MSTID signatures and their association with nighttime SF over Europe. The results highlight that MSTID occurrence serves as a strong indicator of favorable conditions for the development of instabilities in the mid-latitude ionosphere, offering valuable implications for the possible modeling of the ionosphere and the prediction of ionospheric weather. Full article

Corrosion-severity grading of grounding-grid connectors from optical images supports proactive power-infrastructure maintenance. Existing approaches rely on single-time-point, manually thresholded hue–saturation–value (HSV) metrics and static multi-criteria decision-making (MCDM) frameworks that cannot capture corrosion dynamics. In this paper we present a pipeline that (1) defines a four-class corrosion grade from an HSV area fraction (S_corr) measured on RGBA optical images, and validates those labels against a baseline-referenced CIEDE2000 metric zero-referenced to each connector’s as-received appearance; (2) generates 240 color-prior-constrained procedural synthetic images from 53 real images across six connector types; (3) fine-tunes a ResNet-18 to estimate corrosion coverage continuously, deriving the reported severity class from that estimate rather than predicting it directly; and (4) fits power-law kinetics C(t) = k·tⁿ to the S_corr time series, propagates bootstrap uncertainty into a Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) framework, and reports kinetics-aware rankings as rank probabilities. The label validation quantifies two limitations of single-threshold HSV grading: a material-color offset that scores an unexposed copper connector at S_corr = 0.442, and insensitivity to achromatic corrosion products covering roughly 80% of the aluminum and galvanized-steel surface. Ablation experiments replicated over five random seeds show that neither contribution claimed from a single run survives replication: synthetic augmentation changes macro-F1 by +0.050 (p = 0.46) under the adopted checkpoint-selection rule and by −0.059 (p = 0.43) under the rule used in the original experiments, and the monotonicity-consistency loss by −0.011 (p = 0.87) and +0.001 (p = 0.99) respectively; the previously reported single-run values of 0.208 and 0.494 are draws from opposite tails of the same seed distributions (0.403 ± 0.140 and 0.344 ± 0.073). The one formulation that improves significantly is the continuous one adopted here, which raises Spearman agreement with the independent metric from 0.316 ± 0.150 to 0.698 ± 0.108 (p = 0.005). Measured against controls, a classifier that never sees the image reaches macro-F1 = 0.425 and, after Holm–Bonferroni correction, no deep configuration is distinguishable from it; none exceeds a one-dimensional linear rule on S_corr (0.664); and under leave-one-material-out cross-validation the network does not improve on S_corr used directly as a predictor (ρ = +0.627 against +0.744, paired p = 0.14). Time-resolved energy-dispersive X-ray spectroscopy (EDS) provides a partial chemical consistency check, with welding at ρ = 0.82 (raw p = 0.023), but no material survives Holm correction across the six tested. A U-Net segmentation head supervised only by synthesis-derived masks attains Dice = 0.85 in-domain and collapses to a 0.033 output range on real images, 5% of the HSV metric’s range; the photometric-stability advantage previously claimed for it is an artifact of that collapse and is withdrawn. Kinetics-aware MCDM with propagated uncertainty resolves 9 of 15 pairwise orderings, placing stainless steel above welding at 30 chamber days with probability 1.000 and reversing the static ranking. The pipeline, code and fixed data split are fully reproducible (random seed 42). Full article

Static forecast models used by gas distribution operators in regulatory submissions provide point estimates but cannot quantify the probability that approved targets will be met. This paper proposes an Extended Kalman Filter–Monte Carlo (EKF–MC) framework that (i) jointly estimates the residential client state and the parameters of the logistic S-curve growth model in a sequential Bayesian setting, and (ii) propagates the posterior parameter uncertainty through 100,000 Monte Carlo draws to construct calibrated probability distributions for the 2026–2027 regulatory forecast horizon. Applied to five years of regulatory submission data (2021–2025) for a licensed gas distribution operator in Sofia Province, Bulgaria, the EKF refines the saturation ceiling to M = 1995 ± 30 and the growth rate to r = 0.561 ± 0.083. The Monte Carlo analysis yields 90% forecast intervals of [1930–2002] for 2026 and [1939–2015] for 2027. Both intervals lie entirely below the regulator-approved targets (2039 and 2150), demonstrating a structural over-forecasting tendency in the regulatory approval process. The framework provides operators with a computationally efficient, auditable tool for quantifying forecast uncertainty in rate-case submissions and capital investment planning. Full article

Existing runway incursion risk approaches can estimate risk under specified conditions but have limited ability to represent how runway protection constraints weaken across operational stages. This study develops an integrated System-Theoretic Process Analysis (STPA)–Dynamic Bayesian Network (DBN) framework to examine how safety-constraint degradation propagates across predefined operational stages toward runway incursion risk. A Dynamic Causal Translation Logic (DCTL) provides principle-guided organization of STPA-derived artifacts into a traceable single-slice Bayesian Network (BN), which is subsequently extended across semantic stages through selected inter-slice transitions and sequential evidence updating. The model was parameterized using 66 manually coded runway incursion-related reports from the Aviation Safety Reporting System (ASRS). Intra-slice conditional probability tables (CPTs) were estimated and calibrated from these reports, while transition CPTs used expert-informed first-order Markov assumptions. The 2023 runway incursion event at John F. Kennedy International Airport (JFK) was used as a single case-informed demonstration of staged inference and does not constitute independent or cross-case validation. Posterior updating indicated weakened unobserved monitoring and verification constraints, higher unsafe control action likelihood, and sustained tendencies toward runway occupancy conflict and Category B severity. Scenario-based analysis showed that alternative safety-constraint states produced distinct local changes in downstream hazard and Category B posterior probabilities under their respective stage-specific conditioning settings. The framework supports tracing safety-constraint degradation and stage-specific runway safety management. Full article

Unmanned Underwater Vehicle (UUV) swarms operating in complex marine environments must accurately assess threats from surrounding targets to ensure mission success and navigational safety. However, existing threat assessment methods face three fundamental bottlenecks when applied to underwater swarms: the inability to track temporally evolving target intentions, reliance on subjective indicator weighting for multi-target ranking, and vulnerability to spatially heterogeneous sonar noise. This paper proposes a hierarchical threat assessment framework that addresses these bottlenecks through three integrated modules. First, a Dynamic Bayesian Network with a specially designed heading factor tracks target intention over time, propagating threat probabilities across sequential observations and enabling early warning before the closest point of approach. Second, a Vieta’s theorem-based algebraic ranking algorithm constructs comprehensive threat vectors via elementary symmetric polynomials of six indicator utilities, avoiding explicit expert-defined weighting coefficients in the multi-attribute ranking stage while capturing both independent and synergistic indicator interactions. Third, a distance-weighted swarm aggregation strategy suppresses individual sonar noise by assigning higher fusion weights to geographically closer nodes, exploiting the spatial diversity inherent in swarm configurations. Simulation experiments under representative target-motion scenarios validate the framework across four complementary experimental studies. Results demonstrate that the DBN reduces output variance by over 56% compared to static Bayesian networks and responds to abrupt intention changes within 15 s. The algebraic ranking algorithm achieves identical prioritization to TOPSIS without requiring any manual or data-dependent weights. The distance-weighted aggregation reduces root mean square error by 63.2% and improves signal-to-noise ratio by 8.7 dB over equal-weight averaging. The proposed framework provides a principled and interpretable solution for simulation-based autonomous threat perception in representative underwater swarm scenarios. Full article

We study prediction-to-decision certification for inverse design with vector-valued response surfaces. An unknown response map is estimated from data, a target response is prescribed, and a decision is obtained by minimizing a target-loss function. The main question is how statistical prediction error and approximate global optimization error propagate to the true decision quality. We prove an end-to-end certificate showing that a high-probability uniform response bound and a certified global-search tolerance imply a high-probability bound on the true excess risk of the selected decision. Under a growth condition, the same event also yields an explicit distance-to-argmin bound. We provide finite-sample ordinary least squares response certificates, conditional ridge certificates with explicit bias decomposition, certified Lipschitz branch-and-bound, and polynomial sum-of-squares formulations. The formal certificate is demonstrated on a controlled synthetic benchmark. A clay-coloration example illustrates the workflow, while the pilot measurements are treated only as local forward-color checks. Full article