Search Results (3,503)

In this paper, we research multi-robot coverage path planning (MCPP), which generates paths for agents to visit all target areas or points. This problem is common in various fields, such as agriculture, rescue, 3D scanning, and data collection. Algorithms to solve MCPP are generally categorized into online and offline methods. Online methods work in an unknown area, while offline methods generate a path for the known. Recently, offline MCPP has been researched through various approaches, such as graph clustering, DARP, genetic algorithms, and deep learning models. However, many previous algorithms can only be applied on grid-like environments. Therefore, this study introduces an offline MCPP algorithm that applies graph-adapted K-means and spanning tree coverage for robust operation in non-grid-structure maps such as road networks. To achieve this, we modify a cost function based on the travel distance by adjusting the referenced clustering algorithm. Moreover, we apply bipartite graph matching to reflect the initial positions of agents. We also introduce a cluster-level graph to alleviate local minima during clustering updates. We compare the proposed algorithm with existing methods in a grid environment to validate its stability, and evaluation on a road network digital twin validates its robustness across most environments. Full article

The proliferation of wireless sensor networks in remote and inaccessible areas demands efficient data collection approaches that minimize energy consumption while ensuring comprehensive coverage. Traditional data retrieval methods face significant challenges when sensors are sparsely distributed across extensive areas, particularly in scenarios where direct sensor access is impractical due to terrain constraints or operational limitations. This research addresses these challenges through a novel hybrid optimization framework that combines integer linear programming (ILP) with multiple traveling salesperson problem (mTSP) algorithms for drone-based data collection in wireless sensor networks (WSNs). The methodology employs a two-phase approach, where ILP optimally determines strategic access point locations for sensor clustering based on communication capabilities, followed by mTSP optimization to generate efficient inter-AP flight trajectories rather than individual sensor visits. Comprehensive simulations across diverse network configurations and drone quantities demonstrate consistent performance improvements, with travel distance reductions reaching 32% compared to conventional mTSP implementations. Comparative evaluation against established clustering algorithms including Voronoi, DBSCAN, Constrained K-Means, Graph-Based clustering, and Greedy Circle Packing confirms that ILP consistently achieves optimal access point allocation while maintaining superior routing efficiency. Additionally, a novel quality assessment metric quantifies sensor grouping effectiveness, revealing that ILP-based clustering advantages become increasingly pronounced with higher sensor densities, providing substantial operational benefits for large-scale wireless sensor network deployments. Full article

The passenger car market has experienced a radical shift: the rise of SUV, crossover vehicles, but also Battery Electric Vehicle (BEV) and Plug-In Hybrid Vehicle (PHEV), has blurred the borders between traditional vehicle segments as well as body types, resulting in reduced applicability of conventional taxonomies of vehicle types. This study aims to provide an overview of the vehicle market by proposing a new, machine-learning-based segmentation of the entire German vehicle fleet covering the past years. We merge over 40 million registered vehicles with a technical specifications database and apply data-mining techniques to derive an improved market segmentation. We demonstrate that unsupervised learning techniques, specifically Ward and k-means clustering, yield clusters with enhanced separation, clarity, and practical usability. Clustering was applied to both raw technical features and engineered features designed to capture aspects of economy, ecology, usability, and performance. The silhouette scores can reach 0.19, a significant increase over the +0.05/−0.05 scores of the existing vehicle segments or chassis types. Full article

This paper presents a real-time terrain mapping method for aircraft’s navigation, combining probabilistic terrain modeling with adaptive azimuth scan command adjustment. The method refines a preloaded DTED in real time using radar scan data, enabling aircraft to update and utilize terrain elevation information during flight. The terrain is represented using a Gaussian Mixture Model (GMM), where radar scan data are evaluated based on their posterior responsibilities. A conditional nested GMM refinement is selectively applied in structurally ambiguous regions to capture multi-modal elevation patterns. The azimuth scan command is adaptively adjusted based on posterior responsibilities by increasing the step size in well-mapped regions and decreasing it in areas with low responsibility. This lightweight and adaptive strategy supports real-time operation with low computational cost. Simulations across diverse terrain types demonstrate accurate grid updates and adaptive scan control, with the proposed method achieving max error 29 m compared to grid-based averaging of 43 m and K-means clustering of 81 m. As the total number of updates is comparable to the existing methods, the proposed approach offers an advantage for real-time applications with enhanced grid accuracy. Full article

In this study, we provide, to our knowledge, one of the first multidimensional, data-driven evaluations of green education performance in Turkish higher education, combining the THE Education Score, THE Impact Score, and the UI GreenMetric Education & Research Score (GM-ED) with institutional characteristics, and situating the analysis within SDG 4 (Quality Education). While universities worldwide increasingly integrate sustainability into their missions, systematic evidence from middle-income systems remains scarce. To address this gap, we compile a dataset of 50 Turkish universities combining three global indicators—the Times Higher Education (THE) Education Score, THE Impact Score, and the UI GreenMetric Education & Research Score (GM-ED)—with institutional characteristics such as ownership and student enrollment. We employ descriptive statistics; correlation analysis; robust regression models; composite indices under equal, PCA, and entropy-based weighting; and exploratory k-means clustering. Results show that integration of sustainability into curricula and research is the most consistent predictor of SDG-oriented performance, while institutional size and ownership exert limited influence. In addition, we propose composite indices (GECIs). GECIs confirm stable top performers across methods, but mid-ranked universities are volatile, indicating that governance and strategic orientation matter more than structural capacity. The study contributes to international debates by framing green education as both a measurable indicator and a transformative institutional practice. For Türkiye, our findings highlight the need to move beyond symbolic initiatives toward systemic reforms that link accreditation, funding, and governance with green education outcomes. More broadly, we demonstrate how universities in middle-income contexts can institutionalize sustainability and provide a replicable framework for assessing progress toward SDG 4. Full article

Fine-grained sedimentary rocks exhibit significant textural heterogeneity, often obscured by conventional grain size analysis techniques that require sample disaggregation. We propose a non-destructive, image-based grain size characterization workflow, utilizing stitched polarized thin-section photomicrographs, k-means clustering, and watershed segmentation algorithms. Validation against laser granulometry data indicates strong methodological reliability (absolute errors ranging from −5% to 3%), especially for particle sizes greater than 0.039 mm. The methodology reveals substantial internal heterogeneity within Es3 laminated shale samples from the Shahejie Formation (Bohai Bay Basin), distinctly identifying coarser siliceous laminae (grain size >0.039 mm, Φ < 8 based on Udden-Wentworth classification) indicative of high-energy depositional environments, and finer-grained clay-rich laminae (grain size <0.039 mm, Φ > 8) representing low-energy conditions. Conversely, massive mudstones exhibit comparatively homogeneous grain size distributions. Additionally, a multifractal analysis (Multifractal method) based on the S_50bi/S_50si ratio further quantifies spatial heterogeneity and pore-structure complexity, significantly enhancing facies differentiation and reservoir characterization capabilities. This method significantly improves facies differentiation ability, provides reliable constraints for shale oil reservoir characterization, and has important reference value for the exploration and development of the Bohai Bay Basin and similar petroliferous basins. Full article

A fundamental challenge in lithium-ion battery charging is the inherent trade–off between charging speed and battery health. Fast charging tends to accelerate battery degradation, while slow charging extends downtime and intensifies range anxiety, heightening concerns over inadequate driving range during operation. This contradiction has become a key bottleneck restricting the advancement of electric vehicles. In response to the limitations of conventional charging strategies and optimization methods, which typically intensify this trade–off, this study proposes a novel two–stage fast charging optimization strategy for lithium–ion batteries. The proposed method first introduces a hybrid clustering algorithm that combines the canopy algorithm with bisecting K–means to achieve adaptive SOC staging. This staging is guided by the nonlinear characteristics of the internal resistance with respect to the state of charge (SOC), allowing for a data–driven division of charging phases. Following staging, a closed–loop optimization framework is developed. A wavelet neural network (WNN) is employed to precisely capture and approximate the nonlinear characteristics of the charging process for performance prediction, upon which a multi–strategy enhanced multi–objective particle swarm optimization (MOPSO) algorithm is applied to efficiently search for Pareto–optimal solutions that balance charging time and ohmic loss. In addition, an active learning mechanism is incorporated to refine the WNN using selectively sampled data iteratively, thereby improving prediction accuracy and the robustness of the optimization process. Experimental results demonstrate that when the SOC reaches 70%, the proposed method shortens the charging time by 12.5% and reduces ohmic loss by 31% compared with the conventional constant current–constant voltage (CC–CV) strategy, effectively achieving a balance between charging efficiency and battery health. Full article

Oxidative stress is a key driver of endothelial dysfunction and early cardiovascular risk. Antioxidant enzymes such as superoxide dismutase (SOD) and glutathione peroxidase (GPX) are vital for vascular protection, especially during growth. While exercise-induced redox adaptations are well established in adults, data in pediatric athletes are limited. This cross-sectional study examined associations between training load and systemic antioxidant enzyme activity in 203 youth athletes aged 10–16 years, also considering sex, age, sports discipline, and redox phenotypes. Physical activity was assessed via validated questionnaires and expressed as weekly hours and MET-hours. Fasting blood samples were analyzed for SOD and GPX. Statistical tests included t-test, ANOVA, regression, and k-means clustering. Antioxidant enzyme levels were stable across training volumes, sports disciplines, and age groups. Boys showed significantly higher SOD than girls (259.43 ± 54.02 U/mL vs. 226.93 ± 48.22 U/mL, p < 0.001); GPX levels were similar between sexes. Cluster analysis identified three distinct redox profiles with differing training and sex distributions. No linear association was observed between training load and enzyme activity. Findings suggest that youth athletes exhibit robust antioxidant defenses, with individual and sex-related factors playing a more prominent role than training volume. These results highlight the value of regular physical activity for vascular health during development and the need for longitudinal studies to track redox adaptation. Full article

(1) Background: University students face increasing mental health challenges, with sociodemographic disparities shaping well-being outcomes and highlighting the need for machine learning approaches to identify distinct psychological profiles. (2) Methods: This cross-sectional study surveyed 4911 Chinese university students (aged 18–25) using the EPOCH Questionnaire, which measures Engagement, Perseverance, Optimism, Connectedness, and Happiness. Data were collected via WenjuanXing (WJX), with recruitment promoted through official channels. Well-being profiles were identified through exploratory K-means clustering, with internal validity and the optimal cluster number assessed using the silhouette coefficient. (3) Results: Cluster analysis identified two distinct groups: Cluster 0 (41.09%) with higher well-being scores and Cluster 1 (58.91%) with lower scores. Differences across all five EPOCH dimensions exceeded 1.0, most notably in Optimism (Δ = 1.31) and Happiness (Δ = 1.37). A subgroup of concern within Cluster 1 (n = 92), primarily male sophomores from rural, low-income, multi-child families receiving financial aid, showed particularly low scores in Connectedness (Δ = −0.57) and Happiness (Δ = −0.43). In contrast, a high well-being subgroup in Cluster 0 (n = 108), mainly urban female freshmen from high-income, only-child families, exhibited elevated scores, especially in Connectedness (Δ = 0.69) and Happiness (Δ = 0.65). (4) Conclusions: This exploratory clustering study identified distinct well-being profiles among Chinese university students, with demographic and socioeconomic vulnerabilities associated with diminished psychological well-being, particularly in Connectedness, Happiness, and Optimism. These findings highlight the need for targeted interventions that integrate psychosocial support with financial assistance to reduce inequalities and promote flourishing. Full article

The increasing global demand for renewable energy poses significant challenges to grid stability due to the fluctuation and unpredictability of photovoltaic (PV) power generation. To enhance the accuracy of short-term PV power prediction, this study proposes an innovative integrated model that combines Convolutional Neural Networks (CNN) and Bidirectional Long Short-Term Memory (BiLSTM), optimized using the Starfish Optimization Algorithm (SFOA) and integrated with a multi-method data processing framework. To reduce input feature redundancy and improve prediction accuracy under different conditions, the K-means clustering algorithm is employed to classify past data into three typical weather scenarios. Empirical Mode Decomposition is utilized for multi-scale feature extraction, while Kernel Principal Component Analysis is applied to reduce data redundancy by extracting nonlinear principal components. A hybrid CNN-BiLSTM neural network is then constructed, with its hyperparameters optimized using SFOA to enhance feature extraction and sequence modeling capabilities. The experiments were carried out with historical data from a Chinese PV power station, and the results were compared with other existing prediction models. The results demonstrate that the Root Mean Square Error of PV power generation prediction for three scenarios are 9.8212, 12.4448, and 6.2017, respectively, outperforming all other comparative models. Full article

The dynamic development of agent systems and large language models opens up new possibilities for automating spatial and investment analyses. The study evaluated a reactive AI agent with an NLP interface, integrating Apache Spark for large-scale data processing with PostGIS as a reference point. The analyses were carried out for two areas: Nowy Sącz (36,000 plots, 7 layers) and Ostrołęka (220,000 plots). For medium-sized datasets, both technologies produced similar results, but with large datasets, PostGIS exceeded time limits and was prone to failures. Spark maintained stable performance, analyzing 220,000 plots in approximately 240 s, confirming its suitability for interactive applications. In addition, clustering and spatial search algorithms were compared. The basic DFS required 530 s, while the improved one reduced the time almost tenfold to 54–62 s. The improved K-Means improved the spatial compactness of clusters (0.61–0.76 vs. <0.50 in most base cases) with a time of 56–64 s. Agglomerative clustering, although accurate, was too slow (3000–6000 s). The results show that the combination of Spark, improved algorithms, and agent systems with NLP significantly speeds up the selection of plots for renewable energy sources, supporting sustainable investment decisions. Full article

The university period is a transitional stage during which students develop heterogeneous interpersonal styles to navigate complex social demands. While prior studies have linked interpersonal functioning to adaptability and mental health, structural differences across interpersonal style groups remain underexplored. Therefore, the current research was designed to examine whether and how adaptability is differentially related to mental health symptoms when considered within the framework of distinct interpersonal style profiles. Using K-means clustering, we identified three distinct interpersonal profiles: the withdrawn and avoidant type, the overinvolved and compliant type, and the well-adjusted interpersonal type. Based on this classification, network analyses were conducted to examine how six dimensions of adaptability related to three core mental health symptoms within each group. The results showed a consistent pattern across all profiles, with emotional adaptability negatively associated with depression, anxiety, and stress. Subsequent network comparison analyses demonstrated that the withdrawn and avoidant group differed significantly in structure from the well-adjusted interpersonal group, particularly in the connections involving emotional, interpersonal, and economic adaptability. By uncovering meaningful differences in adaptability-mental health associations across interpersonal style, this study provides a foundation for designing targeted strategies that address the unique adaptabilities and mental health problems of distinct interpersonal profiles. Full article

Background/Objectives: Type 2 diabetes (T2D) is a clinically and genetically heterogeneous disease. In this study, we aimed to stratify patients with T2D from the Volga-Ural region of Eurasia into distinct subgroups based on clinical characteristics and to investigate the genetic underpinnings of these clusters. Methods: A total of 254 Tatar individuals with T2D and 361 ethnically matched controls were recruited. Clinical clustering was performed using k-means and hierarchical algorithms on five variables: age at diagnosis, body mass index (BMI), glycated hemoglobin (HbA1c), insulin resistance (HOMA-IR), and β-cell function (HOMA-B). Genetic association analysis was conducted using logistic regression under an additive model, adjusted for age and sex, and corrected for multiple comparisons using the Benjamini–Hochberg method. Results: Four distinct T2D subtypes were identified—mild age-related diabetes (MARD, n = 25), mild obesity-related diabetes (MOD, n = 72), severe insulin-resistant diabetes (SIRD, n = 66), and severe insulin-deficient diabetes (SIDD, n = 52)—each with unique clinical and comorbidity profiles. SIDD patients exhibited the highest burden of microvascular complications and lowest estimated glomerular filtration rate. Nine genetic variants showed significant associations with T2D and/or specific subtypes, including loci in genes related to neurotransmission (e.g., HTR1B, CHRM5), appetite regulation (NPY2R), insulin signaling (TCF7L2, PTEN), and other metabolic pathways. Some variants demonstrated subtype-specific associations, underscoring the genetic heterogeneity of T2D. Conclusions: Our findings support the utility of clinical clustering in uncovering biologically and clinically meaningful T2D subtypes and reveal genetic variants that may contribute to this heterogeneity. These insights may inform future precision medicine approaches for T2D diagnosis and management. Full article

Lithium-ion batteries (LIBs) are critical components in safety-critical systems such as electric vehicles, aerospace, and grid-scale energy storage. Their degradation over time can lead to catastrophic failures, including thermal runaway and uncontrolled combustion, posing severe threats to human safety and infrastructure. Developing a robust AI framework for degradation prognostics in safety-critical systems is essential to mitigate these risks and ensure operational safety. However, sensor noise, dynamic operating conditions, and the multi-scale nature of degradation processes complicate this task. Traditional denoising and modeling approaches often fail to preserve informative temporal features or capture both abrupt fluctuations and long-term trends simultaneously. To address these limitations, this paper proposes a hybrid data-driven framework that combines Sample Entropy-guided Variational Mode Decomposition (SE-VMD) with K-means clustering for adaptive signal preprocessing. The SE-VMD algorithm automatically determines the optimal number of decomposition modes, while K-means separates high- and low-frequency components, enabling robust feature extraction. A dual-branch architecture is designed, where Gated Recurrent Units (GRUs) extract short-term dynamics from high-frequency signals, and Transformers model long-term trends from low-frequency signals. This dual-branch approach ensures comprehensive multi-scale degradation feature learning. Additionally, experiments with varying sliding window sizes are conducted to optimize temporal modeling and enhance the framework’s robustness and generalization. Benchmark dataset evaluations demonstrate that the proposed method outperforms traditional approaches in prediction accuracy and stability under diverse conditions. The framework directly contributes to Artificial Intelligence for Security by providing a reliable solution for battery health monitoring in safety-critical applications, enabling early risk mitigation and ensuring operational safety in real-world scenarios. Full article

Background and Objectives: Stroke is widely recognised for its physical consequences. However, cognitive and emotional impairments, such as depression, anxiety, and vascular cognitive impairment (VCI), are often under-recognised and under-treated. Our study aimed to identify and characterise cognitive and emotional sequelae in patients hospitalised for acute ischemic stroke. Materials and Methods: We conducted a cross-sectional study involving 73 patients within seven days of an acute ischemic stroke. Patients were assessed using the National Institutes of Health Stroke Scale (NIHSS), modified Rankin Scale (mRS), Montreal Cognitive Assessment (MoCA), Hachinski Ischemic Score (HIS), and the Clinical Assessment of Depression (CAD) questionnaire, which includes four subscales (Depressed Mood (DM), Anxiety/Worry, Disinterest, and Physical Fatigue). K-means clustering was applied to ten standardised clinical and psychometric variables. In addition, multiple linear regression was performed to determine independent predictors of cognitive and affective outcomes, with MoCA and CAD-DM as dependent variables. Results: Three distinct patient profiles emerged: (1) Mild Impairment Profile, characterised by minimal cognitive or emotional symptoms; (2) Depressive Profile, marked by elevated emotional symptom scores despite mild physical impairment; and (3) Vascular Cognitive Impairment Profile, comprising older patients with the most severe cognitive and functional deficits. ANOVA confirmed significant differences between groups in NIHSS, mRS, MoCA, HIS, and CAD scores, but not for age or education. Linear regression revealed that older age (β = –0.10, p = 0.012) and higher NIHSS at discharge (β = –0.72, p = 0.020) predicted lower MoCA scores, whereas years of education (β = 0.58, p = 0.013) predicted better cognition (R² = 0.29). No demographic or clinical factors predicted depressive symptoms (all p > 0.29). Conclusions: Our study highlights the heterogeneity of post-stroke outcomes. Neuropsychiatric impairments may be present even in patients with minimal physical deficits and require targeted evaluation and management. Full article