Search Results (422)

Aiming at the problem that complex underwater environments induce outliers in Doppler Velocity Log (DVL) measurements, which degrade the navigation accuracy of the Strapdown Inertial Navigation System (SINS)/DVL integrated system, this paper proposes a soft-gating Gaussian–Student’s t mixture robust Kalman filter (MRKF). Firstly, the measurement noise is modeled as a mixture of Gaussian and Student’s t distributions to adapt to normal stationary noise and abrupt outliers, respectively. Secondly, a logistic soft-gating weight is constructed based on the innovation Mahalanobis distance to adaptively balance the output contributions of the standard Kalman Filter (KF) and the variational Bayesian Student’s t filter. Finally, moment matching is adopted to realize the weighted fusion of two-branch posterior distributions, and an equivalent Gaussian posterior estimation is obtained. Simulation results under the considered SINS/DVL integrated navigation scenarios show that the proposed MRKF maintains estimation accuracy close to the standard KF under nominal Gaussian measurement noise. In the designed DVL outlier-injection scenario, the proposed MRKF achieves a position RMSE of $53.39\mathrm{m}$, compared with $878.75\mathrm{m}$, $58.84\mathrm{m}$, and $56.49\mathrm{m}$ for the nominal KF, Huber KF (HKF), and Student’s-t variational Bayesian KF (STVBKF), respectively. These results indicate that the proposed MRKF can improve robustness against DVL outliers while maintaining competitive estimation accuracy under the simulated conditions. Full article

Terrestrial ecosystems are facing increasingly severe threats driven by the dual pressures of climate change and anthropogenic activities. However, current remote sensing-based ecological research still exhibits notable deficiencies in the integration of multi-source data. This study develops a Critical Transition Index (CTI) for Northeast China. The CTI integrates four remotely sensed vegetation variables (LAI, NDVI, SIF, and VOD) with time series decomposition (STL), multiple early-warning signals (ar1, variance, skewness, and kurtosis), consistency scoring, and Mahalanobis distance. The framework systematically assesses vegetation resilience and its spatiotemporal responses to climatic stressors. Results reveal pronounced differences among variables: the structural indicator LAI identified the highest proportion of high-risk areas (60.8%, CTI ≥ 0.8), whereas the functional indicator SIF showed relatively high stability, with a mean CTI of 0.619 and a high-risk proportion of only 16.0%. High-risk areas are primarily concentrated in cropland–grassland mosaics, while forested regions maintain lower risk. Temporal analysis of land cover composition within high-risk areas shows a clear “structural diffusion” trend: the proportion of deciduous broadleaf forests in the high-risk category increased from being negligible in early periods (2003–2007) to approximately 20% in later periods (2013–2017) for both SIF and VOD indicators. This study underscores the necessity of multi-indicator frameworks for detecting critical transitions and provides quantitative, spatially explicit scientific insights for ecosystem early-warning and regional management strategies. Full article

Optimizing speed breeding protocols for biennial crops requires matching the vernalization regime with the genetic background. In this study, nine sugar beet genotypes were exposed to 12, 13, 14 or 15 weeks of vernalization and subsequently grown under controlled speed breeding conditions. Survival analysis revealed a threshold-like acceleration of bolting and flowering: 12 and 13 weeks were largely equivalent, whereas 14–15 weeks sharply increased the bolting and flowering hazard rates. Genotypic variation strongly influenced reproductive success and seed yield traits; genotype MARGARITA KWS combined early flowering with the highest seed number (361 seeds per plant) and total seed weight (5.26 g), while genotype 1K073 did not flower under any vernalization duration. A separate mini-steckling root architecture experiment with 11 genotypes showed that slow-release Osmocote fertilizer significantly increased mini-steckling fresh weight, length and width, with the strongest responses in genotypes 1K073, 1K139 and SMART LIENNA KWS. The interaction between genotype and nutrition was significant for mini-steckling fresh weight and width, indicating that optimal nutrition can modulate the expression of genotypic differences. Multivariate analyses (PCA, CVA, Mahalanobis distances) confirmed that vernalization duration had a threshold-type effect and that genotype was the dominant factor for seed traits, whereas nutrition was the main driver of mini-steckling architecture. Overall, these findings suggest that tailoring vernalization duration and nutrition to the genetic background may substantially improve the efficiency of sugar beet speed breeding. Full article

Autoencoders (AEs) have been widely used for industrial process fault detection owing to their ability to learn nonlinear representations from normal operating data. However, conventional AE methods rely heavily on reconstruction errors and may miss weak faults due to overgeneralization. In addition, insufficient modeling of temporal evolution and operating condition variations may reduce their sensitivity to dynamic faults. To address these issues, this study proposes a memory-enhanced and prediction-assisted conditional variational autoencoder named MI-CVAE for unsupervised fault detection. In the proposed framework, statistical features extracted from sliding windows are used as condition information to describe variable operating states. A memory module stores representative normal prototypes to constrain reconstruction and reduce overgeneralization to faulty samples. Meanwhile, an Informer branch captures temporal dependencies and provides complementary prediction residuals. Reconstruction and prediction residuals are fused to construct squared prediction error and squared Mahalanobis distance statistics, with control limits determined by kernel density estimation. The proposed method is validated on the Benchmark Simulation Model No. 1 wastewater treatment benchmark and a real papermaking process dataset. The results show that MI-CVAE outperforms the evaluated comparison methods, particularly in detecting weak and dynamic faults, while maintaining a low false alarm rate. Full article

The efficiency of converting ripe fruits into processed beans is an economically relevant component of Coffea canephora production systems, yet its genetic parameters remain poorly characterized in studies that do not partition the genotype × year (G×Y) interaction. This study estimated genetic parameters for five processing efficiency traits, namely grain proportion (% grain), husk proportion (% husk), fruit fresh mass per grain mass (FWM/GW), fruit fresh mass per bag (FWM/bag), and fruit volume per bag (FVol/bag), in 48 C. canephora genotypes (40 Robusta, 8 Conilon) evaluated over two crop years (2023–2024) in Jaguaré, Espírito Santo, Brazil. Bayesian inference via MCMC (brms) revealed that the genotype × year variance component exceeded the genotypic variance in 79–97% of posterior samples across the 48 genotypes evaluated over two crop years, a result that should be interpreted within the context of this restricted temporal window, with median heritabilities of 0.27–0.50 (95% credible intervals spanning up to 0.66 units, reflecting the uncertainty inherent to the two-year evaluation window) and genotypic correlations of 0.19–0.38 between years, indicating low consistency of genetic merit across crop seasons. Bayesian probability of consistent superiority identified Z21 as the genotype with the highest predictability for FWM/bag (prob. $=0.846$ at 20% selection intensity), while VR3 showed a favorable profile across four traits simultaneously. The multi-trait model with unstructured covariance estimated a negative genetic correlation between % grain and FWM/bag (${\widehat{r}}_g=-0.87$), suggesting potential for indirect selection. UPGMA clustering based on Mahalanobis distance (CCC $=0.813$) yielded six divergence groups that did not coincide with the botanical classification Conilon/Robusta. In this single-location, two-year study, temporal instability was the predominant source of uncertainty in the selection for processing efficiency in C. canephora under restricted evaluation windows. under restricted evaluation windows. Accordingly, the highlighted genotypes should be interpreted as priority candidates for validation in multi-environment, multi-year networks, rather than as definitive cultivar recommendations, given that the short evaluation window limits the generalizability of genotypic rankings. Full article

Tomato is an important vegetable crop suited to both open-field and protected cultivation. Indeterminate genotypes with high yield potential and desirable quality traits are especially suited to off-season production under protected cultivation. The present study evaluated 57 indeterminate tomato genotypes over two consecutive years under protected conditions to assess genetic variability, genetic divergence, and trait associations across 16 important yield-attributing and quality traits. The analysis of variance depicted significant differences among genotypes for all traits under study. The traits, viz., fruit weight and number of fruits per cluster, exhibited high heritability and high genetic gain, suggesting the predominance of additive gene action and the possibility of direct selection. A significant, positive correlation between fruit weight and the number of plant clusters and yield was observed. Analysis of genetic divergence following Mahalanobis D² statistics classified the genotypes into seven clusters. The number of flowers per cluster and fruit width were the top contributors to the total genetic divergence. Cluster VI outperformed for earliness and yield, Cluster V outperformed for nutritional quality, while Cluster VII was superior for fruit size. Principal Component Analysis revealed that the first five components cumulatively explained 83.3% of the total variation, with PC1 defined by fruit number trait and PC2 by yield and earliness traits. The Multi-Trait Genotype-Ideotype Distance Index (MGIDI) was used to select the best-performing genotypes, highlighting PIDGT-39, PIDGT-42, and PIDGT-29 as elite. Thus, the findings of the present study provide deeper insights into the genetic makeup of indeterminate tomato genotypes and potential parental accessions for tomato improvement, to enhance yield and quality under protected conditions. Full article

This study introduces a multivariate distance-based framework for analyzing hospital liquidity stress using three financial indicators: cash reserves, days with negative cash, and accounts receivable. Using Definitive Healthcare data from 2020–2025, the study applies principal component analysis (PCA), Mahalanobis distance, Aitchison distance, and ternary plots to characterize structural relationships among these liquidity variables. The results show that the first two principal components explain more than 94% of the variation in the transformed variables, indicating that the joint financial structure can be represented in a lower-dimensional space. Beginning in 2023, accounts receivable became more geometrically separated from the cash-based variables, suggesting that revenue-cycle dynamics may have become a more independent dimension of hospital liquidity stress. Importantly, this manuscript does not directly predict hospital closure or bankruptcy because verified event/non-event outcome data are not available in the analytic file. Instead, its contribution is methodological and exploratory: it demonstrates how distance-based and compositional methods can identify structural liquidity instability and potential early warning signals that warrant further validation with longitudinal closure, bankruptcy, or severe-distress outcomes. Full article

Conventional settlement monitoring techniques are inadequate for seawall construction environments due to severe physical impacts, the absence of terrestrial communication networks, and highly dynamic disturbances. This research proposes a multi-source fusion settlement monitoring system designed specifically for the construction phase to overcome these constraints. An integrated inclinometer–magnetoresistive sensing unit is the central component of this system. The unit achieves physical isolation from the severe impact loads of rock backfilling, guarantees protection in high-salinity and high-humidity environments, and accommodates the large deformations typical of soft foundations by utilizing a structural design that includes a rigid channel steel sheath, anti-corrosion sealing, and flexible joints. In terms of computation, a cascaded attitude fusion framework is developed that combines a Multiplicative Extended Kalman Filter (MEKF) with Quaternion Estimator (QUEST) initialization. High-precision displacement inversion via quaternion rotation is made possible by the introduction of an adaptive mechanism based on the Mahalanobis distance that precisely detects and suppresses transient acceleration disturbances induced by construction machinery and waves. Additionally, data transmission issues in remote offshore areas are resolved by combining solar power and BeiDou short-message communication technologies. This adaptive technique minimizes attitude estimate errors in dynamic situations by approximately 84.56%, as demonstrated by experimental and field validation. The system was deployed as a 165 m array comprising 49 sensing units and monitored continuously for 458 days, achieving a normalized RMSE of 9.44–11.02% compared to reference settlement tubes and capturing a maximum settlement of 1.7 m in the core high-fill section. These results confirm the system’s high monitoring accuracy and resilience in harsh construction conditions. Full article

Radar technology in the microwave and millimeter-wave frequency range is the subject of current research for structural health monitoring of composite materials, e.g., damage detection in wind turbine blades. Performance assessment, enabling widespread practical application of this promising and non-contact sensing approach, can be realized via probability of detection (POD) theory, which is a statistical method for determining the detectability of damage through response metrics as a function of flaw size. This paper deals with the experimental investigation of a delamination model represented by two parallel glass fiber reinforced polymer plates separated from each other from $0\mathrm{mm}$ to $1\mathrm{mm}$ in steps of $0.01\mathrm{mm}$. Experimental studies with a frequency modulated continuous wave radar are performed under laboratory conditions in the frequency range from $57GHz$ to $65GHz$. The signal response is represented by two damage indicators (DIs), according to the root mean square deviation and Mahalanobis distance. Since the reflection of electromagnetic waves exhibits a nonlinear behavior, this also implies a nonlinear response in the DI characteristic. The novelties in this work are the successful implementation of a nonlinear regression model, combined with an optimal threshold decision through receiver operating characteristic curves for a high-resolution POD representation. The POD with $95\%$ confidence bounds indicates the flaw size at which the delamination can be detected reliably. Depending on the radar distance in experimental studies, the binary structural condition (damaged or undamaged) was correctly assessed from $95\%$ to $100\%$. The minimum detectable size ranges from $0.01\mathrm{mm}$ to $0.08\mathrm{mm}$. Full article

Trustworthy deployment of artificial intelligence in safety-critical systems requires accurate diagnosis of anticipated scenarios and reliable rejection of out-of-distribution (OOD) inputs that fall outside the modeled operational scope. Existing data-driven diagnostic models typically assume that test inputs are drawn from the training distribution or rely on heuristically tuned thresholds that lack enforceable safety guarantees. This article presents SCOPE (Safety-Calibrated Out-of-distribution Prediction via Contrastive Embeddings), a framework integrating supervised contrastive learning with split-conformal prediction to provide statistically grounded OOD rejection with finite-sample false-alarm control. SCOPE employs a causal residual convolutional encoder to map multivariate sensor streams into a hyperspherical embedding space with a compact, class-specific structure. A k-nearest-neighbor density nonconformity score, computed in the encoder embedding space, flags transients that occupy low-density regions relative to known accident manifolds; an ablation shows that this density score outperforms prototype distance, entropy, and conservative maximum fusion as well as a panel of standard OOD baselines (MSP, ODIN, energy, Mahalanobis, OpenMax, MC-dropout, and a reconstruction autoencoder). To support temporally evolving trajectories, SCOPE aggregates window-level scores under a monotone decision policy and performs trajectory-level conformal calibration, yielding distribution-free guarantees that bound the probability of falsely rejecting a known accident run. SCOPE is evaluated on the Nuclear Power Plant Accident Data (NPPAD) benchmark using high-openness splits that withhold entire accident families as unknowns, and all metrics are reported as mean ± standard deviation across multiple random seeds. Results demonstrate strong diagnostic accuracy on accepted trajectories, conservative false-alarm rates satisfying user-specified safety constraints across multiple operating points, and timely rejection of unseen accident mechanisms, making SCOPE suitable for deployment in safety-critical monitoring applications. Full article

The reinforcement and retrofit of damaged steel buildings has emerged as a primary focus in civil engineering. It should be noted that completing the reasonable strengthening design for avoiding the sudden collapse of a structure in extreme engineering conditions was an urgent task, while the existing method required a long time which significantly influenced the reinforcing practice. In the present study, an improved explainable machine learning (ML) framework was developed for the rapid assessment of the bending property of repaired laminated channel beams. Firstly, a comprehensive database of 192 samples combining experimental and finite element data was established. The Mahalanobis distance analysis and Pearson correlation analysis were sequentially performed to evaluate the singularity of the samples and the dependencies between the variables. Secondly, the adversarial tests were conducted on the randomly selected 10 pairs of training and testing sets to determine the database with the best distribution consistency. Then, three machine-learning models of artificial neural networks (ANN), random forest (RF), and extreme gradient boosting tree (XGBoost) were respectively trained and validated. Finally, the explainability analysis of the XGBoost model was carried out in the global and local perspectives based on the SHAP method. The prediction accuracy (R²) of all ML models exceeded 90%, demonstrating good accuracy and providing a useful reference within the current database for the reinforcement design of damaged steel beams in emergency situations. In addition, the XGBoost model achieved superior prediction accuracy (R² = 97.98%) and stability (CoV = 0.82%) compared to ANN and RF. The explainability analysis revealed that boundary conditions and load type had the most significant influence on bending capacity. The proposed ML approach enabled efficient and reliable bending capacity estimation, supporting rapid decision-making in emergency reinforcement scenarios for damaged steel structures. Full article

Indoor service robots often face challenges in target localization and robotic grasping under cluttered backgrounds, partial occlusion, and viewpoint variations. To address these issues, this study proposes a vision-guided robotic grasping framework based on an improved feature matching algorithm termed Mahalanobis-accelerated Scale-Invariant Feature Transform Plus (MSIFT+). The proposed method integrates Mahalanobis distance metric reconstruction with a dynamic Best-Bin-First (BBF) search strategy to improve matching robustness and computational efficiency. A multi-scenario indoor dataset was constructed to evaluate the proposed method under rotational variation, weak-texture, and partial occlusion conditions. The results demonstrate that the MSIFT+ algorithm significantly outperforms other methods in cross-scenario consistency and adaptability to weakly textured targets. Furthermore, a binocular vision-guided robotic grasping system was developed and validated through practical robotic experiments. Experimental results confirm that the MSIFT+ algorithm enhances detection performance for small and clustered targets in complex environments. The proposed framework provides an effective and reliable solution for robotic object localization and grasping in complex indoor environments. Full article

Tropical forest restoration has increased in the past decades, with possible advancements given the UN declaration of the “Decade of Ecosystem Restoration”. However, robust assessments to compare ecosystem functions among restored forest stages are essential. We evaluated 13 actively restored forest stands ranging from 3 to 21 years of age and compared measures of forest biodiversity, structure, and ecosystem function to four 70+ year old “reference” stands that serve as restoration “targets” in the study region of the Premontane wet forest of Costa Rica. The restored stands were planted with an average of 13 tree species on abandoned pastures that were fallow for at least two years. Sixteen tree-stand attributes and six ecosystem function estimates were assessed, including: annual biomass (C) accumulation, N-fixation potential, threatened species conservation, and the provision of avian frugivore forage, insect habitat, and insect pollination. Using Principal Component Analysis, linear modeling, and Mahalanobis distance analyses, we learned that planting a diversity of tree species sets the stage for forest recovery at early restoration ages, with an inflection point at 15 years towards older reference forest characteristics and functions. Given that all restoration ages provided tree diversity and some level of ecosystem functions, the value of all restored stands in the landscape is notable. The assessment methods are easily employed, thereby providing an accessible tool to restoration practitioners. Full article

Rotating machinery is essential in industrial applications, where early fault detection is critical to prevent catastrophic failures. Shafts are mainly vulnerable to imbalances and cracks; these last ones pose a severe risk as they can lead to sudden failure if not identified during their early stages. Cracks induce progressive stiffness reduction, altering the system’s mechanical properties and affecting the forces transmitted to the supports. This study analyses the effects of cracks on a rotating shaft using experimental data. Vibration signals from accelerometers mounted on the supports are processed to identify changes in the shaft’s response. The methodology focuses on distinguishing crack-induced alterations for different imbalance scenarios by analysing key signal features. A statistical detection algorithm and the extracted feature analysis are exploited for crack identification before a critical failure occurs. The results highlight the distinct impact of cracks on the shaft’s dynamic behaviour and demonstrate effective strategies for early detection. While different features highlight the presence of the crack differently, all successfully contribute to detecting the damage. This study provides an analysis of a novel experimental case study for crack detection, enhancing both safety and economic sustainability of rotating machinery. Full article

Near-infrared hyperspectral imaging (NIR-HSI) is widely used as a non-destructive technique for evaluating internal fruit quality; however, reliable pixel-wise visualization remains challenging due to geometry-induced spectral distortions and the lack of statistically interpretable validation criteria. This study proposes an integrated framework for three-dimensional visualization of soluble solids content (SSC) across the entire surface of strawberries using NIR-HSI combined with shape-aware spectral correction and pixel-level reliability assessment. Two complementary imaging systems—a line-scan system and a rotation-scan system—were used to acquire hyperspectral and 3D shape data. Fruit height and surface orientation were incorporated into spectral preprocessing to reduce illumination and curvature effects. Partial least squares regression (PLSR) models were developed using region-of-interest-averaged spectra and applied to pixel-wise SSC mapping. To assess the statistical validity of pixel-level predictions, an imaging reliability index based on the Mahalanobis distance in the PLS score space was introduced. The results show that models with high sample-level accuracy do not necessarily produce reliable SSC maps, whereas reliability-based model selection improves image interpretability. This framework enables consistent three-dimensional SSC visualization and is applicable to hyperspectral imaging of internal fruit attributes. Full article