Topological semantic maps serve as an effective tool for the partially sighted or visually impaired (PSVI) in indoor navigation. Essential to their construction is the precise segmentation of floor plans. However, current techniques fall short in segmentation performance. To overcome this, we introduce a gated-dual-attention-based full-resolution network (GFNet) for floor plan segmentation. We leverage the inherent low-stage detailed features and intraclass-and-interclass contextual dependencies within floor plans to maximize segmentation effectiveness. Our proposed network incorporates modified residual blocks in early stages to maintain full-resolution capture with a low parameter count. We design and apply a novel gated dual attention (GDA) module that efficiently integrates channel and spatial contextual information to enhance local feature representation. This module improves the overall performance of our network while ensuring minimal parameter fluctuation. We also propose a 2-D deep supervision (TDDS) method to merge features from all stages, further enhancing the multilevel feature representation ability. Finally, a practical topological semantic mapping method for PSVI indoor navigation is introduced. All models are evaluated on the rasterized floor plan datasets R2V and R3D. Experimental results show that the proposed network achieves 2.21% and 2.74% mIoU improvements over the previous state-of-the-art method on R2V and R3D, respectively, for the challenging categories of walls and doors. This contributes to more accurate topological semantic mapping.

Urban expansion in areas adjacent to industrial areas presents serious safety challenges, especially in fast developing areas. Nearly between residential and industrial activities increases the risk of accidents, environmental pollution and massive disasters. In response, industrial countries have adopted advanced management systems to reduce these risks. Of these, the Geographical Information System (GIS) provides a powerful tool for analyzing and mapping the areas coming in contact with industrial threats. The purpose of this study is to assess the role of GI in reducing industrial risks and focusing a specific focus on the Algerian context, reducing industrial risks and supporting permanent urban planning. Research adopts a spatial analysis approach using topological maps, geographical datasets and GIS software. Industrial areas and surrounding urban settlements were studied through layered spatial modeling. Supplementary area observation and case studies were included to validate the accuracy of GIS output. The analysis focused on identifying weak areas, assessing industrial threats and assessing the effectiveness of existing land-use and reaction strategies. Conclusions highlight the important weaknesses in current urban risk management systems, especially in terms of preparations and integration with spatial plan. The GIS proved to be effective in detecting high-risk areas, imagining dangerous areas and providing strategic support for the decisions of land-use. The device also featured landscape simulation, which improved the understanding of potential industrial accidents and their impact on nearby communities. The interpretation of these results suggests that GI increases urban flexibility, reduces exposure to industrial threa.

In autonomous exploration tasks in unstructured terrain, exploration efficiency and map topology quality have been a difficult problem to balance. Among the current autonomous exploration methods, geometry-based exploration methods only focus on exploration efficiency but not map quality, which not only leads to frequent backtracking by the robot, but also tends to ignore non-geometric risks such as negative obstacles. To address this pain point, we propose the Structure-Aware Topology Exploration framework. Unlike pure geometric exploration, we utilize U-Net to semantically analyze the unmanned aerial vehicle aerial images, and force the robot’s path to be anchored to the geometric axis of the safe area through the Semantic Seeded Voronoi mechanism. To avoid map redundancy leading to backtracking, we directly introduce topological sparsity constraints in the decision function to realize online structural pruning during exploration. Simulation experiments based on real-world aerial imagery demonstrate that the proposed framework effectively overcomes the late-stage exploration plateau: compared with purely geometric baselines (Rapidly exploring Random Tree and Frontier), it reduces average path length to 278.4 m (45% reduction) and improves exploration efficiency by 80%; compared with the semantic frontier-based baseline, it achieves 28.6% higher efficiency and 13% shorter path length, maximizing information gain per unit travel distance.

Accurate knowledge of Low-Voltage (LV) distribution topology is critical for reliable operation, advanced monitoring and large-scale integration of distributed energy resources (DERs). In practice, topology records in GIS/NIS are frequently incomplete or outdated, while field verification remains costly. This paper presents a smart-meter (SM)–driven methodology for LV topology mapping that combines a data-trained surrogate model with physics-inspired sensitivity analysis. A feedforward Deep Neural Network (DNN), trained on historical SM measurements spanning diverse operating conditions (including DER-driven net generation), is used as a model-free digital twin to emulate customer-to-voltage relationships. Virtual Network Discovery Events (NDEs) are then generated by applying controlled perturbations within the surrogate to obtain voltage-response signatures that support topology inference without physical intervention. Phase groups are identified through dimensionality reduction and hierarchical clustering, and customer connectivity is inferred from the similarity structure of the resulting voltage-response signatures. The method is applied independently per feeder, enabling scalable execution across multi-feeder LV networks. Validation on six real feeders from an urban Spanish network demonstrates accurate voltage emulation and high-fidelity phase and topology reconstruction using only existing SM infrastructure. • Network Discovery Events enable adaptive low-voltage topology mapping. • Hybrid AI–physics approach enhances low-voltage Digital Twins. • Validated on real Spanish distribution network with high precision.

Entropy-driven topology mapping framework for robust Bayesian classification

Abstract The pervasive interconnection of heterogeneous routing devices forms the fundamental infrastructure of modern Internet communication, making accurate router vendor identification a critical capability for multiple domains including network topology mapping, intelligent traffic engineering, and proactive cybersecurity defense. While Internet Protocol version 6 (IPv6) has achieved widespread global deployment as the next-generation Internet protocol, the opaque nature of its addressing mechanisms and protocol behaviors has created significant challenges in router attribute detection across IPv6 networks, leaving a crucial gap in network visibility and security analytics. To address this pressing challenge, we present IPv6 Router FingerPrinting (6RFP), an innovative lightweight fingerprinting methodology that establishes a new paradigm for IPv6 router vendor identification by systematically combining two complementary analytical dimensions: (i) comprehensive EUI-64 interface identifier analysis that captures vendor-specific hardware encoding patterns embedded in IPv6 addresses, and (ii) sophisticated IPv6 Identification Field characteristic profiling that reveals distinctive vendor implementations. Through extensive evaluation across diverse network environments, 6RFP demonstrates highly effective detection capabilities, achieving 85.79% accuracy—representing a remarkable 86.01% improvement over current state-of-the-art techniques—while maintaining minimal computational overhead suitable for real-time deployment.

Abstract Given the multisource and uncertainty of tower crane construction safety risks, this study proposes a comprehensive risk analysis method that integrates disaster chain theory (DCT), analytic hierarchy process (AHP), entropy weight method (EWM), and cloud model. Firstly, based on the disaster chain theory, the topology of risk factors is established, and the key risk factors, such as equipment failure, operation error, environmental interference, and management defects in the operation of tower cranes, are systematically identified. An evaluation system consisting of four first-level indicators and 16 s-level indicators was constructed through the topological mapping of the disaster chain, and the analytic hierarchy process-entropy weight method combined weighting method was used to integrate the subjective and objective weights to effectively balance the expert experience and data objectivity. The cloud model theory is introduced to deal with the ambiguity and randomness in risk assessment, and the mapping relationship between the risk level and the cloud characteristic parameters are established to achieve the conversion from qualitative concept to quantitative evaluation. The case application shows that the risk factors with higher weight ranking include: fatigue/deformation of steel structure, and failure of safety devices. This study innovatively combines dynamic weight optimization with uncertainty modeling to provide a visual risk assessment method for tower crane construction companies.

ABSTRACT Phase singularities, manifesting as null intensity in optical speckle patterns, fundamentally constrain the uniformity of holographic reconstruction. A comprehensive understanding of their intrinsic physical properties is thus critical for advancing coherent optical systems, especially in the pursuit of speckle‐free holography. Using a singularity‐tracking methodology, we numerically investigate the evolution of these singularities as they propagate from holographic planes into free space. Our findings reveal two primary morphological classes of singularity trajectories—closed loops and linear paths—both exhibiting identical fractal characteristics. The topology invariance observed across these trajectories provides a robust framework for achieving high‐uniformity speckle‐free holography, either by eliminating singularities at the initial phase plane or by guiding them outside the target pattern. Notably, the fractal dimension of the trajectories ranges from 1.0 to 1.6 and demonstrates a negative correlation with the F ‐number of the hologram, exemplifying an optical analogy to a biased Brownian random walk under an attractive potential. This work offers novel statistical and topological insights into holographic speckle fields, paving the way for speckle‑less reconstruction using highly coherent light.

To solve the problem on storage and retrieval challenges associated with big data of map, this paper presents an image segmentation- reconstruction algorithm. This algorithm consists of four core steps, i.e., Firstly, it incorporates a semantic association mechanism for image segmentation and develops the corresponding segmentation method. Secondly, it determines the maximum remaining storage capacity of each server unit and performs a matching operation with a preset threshold. Thirdly, it establishes a topological mapping to realize the mapping transformation from the segmented original map to the map to be displayed. Finally, the map is reconstructed by using the location information recorded during the storage of features by image segmentation and mapping relationships. Experimental results show that the accuracy of the proposed reconstruction algorithm can reach 94.35%. To compared with existing image reconstruction algorithms, it not only achieves higher accuracy, faster speed, and stronger anti-interference ability, but also has lower information loss. These advantages of this proposed algorithm can offer an efficient framework for managing large-scale map datasets, thus addressing some critical challenges such as high computational overhead, information loss, and system unreliability in geospatial big data applications.

V4, an intermediate visual area in the ventral pathway of the primate visual system, is known to contain neurons selective to visual stimulus attributes of intermediate complexity. Recent studies have shown that macaque V4 is organized into neuronal columns, each tuned to specific natural image features, and topologically arranged across the cortical surface to form functionally specialized domains. Using digital twins of V4 constructed from a large-scale wide-field imaging dataset, we demonstrate that shape- and texture-preferring neurons-previously identified in single-unit studies-are spatially clustered into functional domains. The segregated spatial organization suggests the existence of parallel modules for surface and boundary processing. Unlike artificial neural networks trained for ImageNet classification, which exhibit a strong texture bias, we find that V4 cortical columns and functional domains are more evenly balanced between shape and texture preferences. Finally, we show that computational constraints of feature similarity and retinotopy constraints are necessary and sufficient to explain many observed properties of the organization of the V4 topological map of natural image feature preferences.

Topological semantic map construction and localization for intelligent vehicles in underground parking lots

Inferring the genetic structure at the subpopulation level is crucial for understanding the demographic histories that shape genetic diversity. Among the most widely used approaches are methods based on admixture and structure modeling-named after the respective software tools-which have become standard due to their intuitive, interpretable outputs. In this study, we address a key methodological question: how does the traditional admixture-based decomposition of genetic components in multilocus population data relate to clustering approaches that leverage machine learning, specifically Self-Organizing Maps (SOMs)? We implemented this approach through our custom SOM-based tool, SOMmelier, which enables the portrayal of genetic structure by identifying modules of co-mutated SNPs and arranging them in a topology-aware genetic landscape. Topology-awareness refers to the organization of genetic modules in a two-dimensional map, where their spatial proximity reflects mutual similarity. We applied Admixture and SOMmelier to investigate the population genetics of European grapevine. Based on prior literature, we considered up to six genetic components, which formed a genetic landscape that closely mirrors the geographic expanse of the classical Mediterranean world-from Western Asia through the Caucasus to Western Europe. The resulting topology reflects the dynamic spatial and temporal nature of grapevine domestication and diffusion. We demonstrate that SOMmelier can recover the genetic components identified by Admixture solely through statistical clustering. By integrating the topological structure of SNP co-variation, it offers perspectives on population structure, evolutionary history, and trait associations in grapevine-and has applicability to other species and systems in population genetics.

ABSTRACT This study proposes a heuristic approach for constructing time-scaled topology maps of indoor WiFi networks using only WiFi log data, without relying on access point (AP) coordinates or signal strength measurements. The method extracts time differences (TD) between consecutive AP transitions from remote authentication dial in user service (RADIUS) server logs. It uses the interquartile range (IQR) method to remove outliers caused by delayed handoffs or inconsistent logging intervals. Descriptive statistics – including the mean and the 25th, 50th, 75th, and 95th percentiles – are then applied to summarize travel behavior and construct time-based distance matrices. A modified Dijkstra algorithm limited to two-hop paths is employed to address missing TD values using both shortest-path and average-path estimation strategies. These completed matrices are projected into two-dimensional space using multidimensional scaling (MDS) to generate topology maps that reflect temporal proximities among APs. The proposed approach is validated through experiments conducted on two floors of a university library, with ground truth walking-time data collected through controlled experiments. Evaluation using Procrustes analysis and normalized Frobenius norm reveal that the 95th percentile provides the most accurate and robust representation of spatial structure, effectively capturing delays and transitional lags commonly observed in indoor mobility. Moreover, average-path Dijkstra estimation outperforms the shortest-path approach in maintaining spatial continuity when direct TD data are missing. This framework enables scalable, low-cost indoor mobility analysis. It offers practical applications in indoor navigation, smart facility management, and spatio-temporal behavior modeling in environments where physical layout information is unavailable or impractical.

The gap between academic curriculum content and modern industrial needs is often an obstacle for fresh graduates in the Information Technology field, particularly in the rapidly evolving Artificial Intelligence (AI) sector. This study aims to identify the relationship patterns among technical competencies (hard skills) most demanded by the global industry. The method employed is Association Rule Mining with the Apriori algorithm to discover association rules between skills, and Network Graph Analysis to visualize the topological map of these competencies. The research dataset covers 15,000 AI job vacancies from the 2024-2025 period, analyzed in depth using Support, Confidence, and Lift Ratio evaluation parameters to validate the strength of relationships between items. The results show that Python is the central competency with the highest frequency of occurrence. Strong association rules were found indicating that proficiency in TensorFlow has a high probability of requiring Python proficiency. The Network Graph visualization reveals three main competency clusters: Data Engineering Ecosystem, Deep Learning, and Infrastructure. These findings offer a strategic foundation for aligning curricula with the job market. Focusing on strengthening the identified competency clusters is expected to directly enhance the relevance and work readiness of graduates.

The shopping service robot (SSR) is designed to offer a superior shopping experience through its continuous target following and companion services. However, the SSR encounters difficulties in complex environments, including visual occlusions and spatial constraints in narrow aisles. To address these difficulties, the study proposes a shopping service robot framework that integrates innovative perception, following control, and path planning modules. Firstly, the perception module employs a multi-sensor fusion method for human target tracking, integrating both red-green-blue-depth (RGB-D) camera and wireless electromagnetic (WEM) data by using an extended Kalman filter (EKF). Secondly, the target-following control module employs an omnidirectional constrained control law, which ensures synchronized orientation alignment between the SSR and the human target. Finally, the path planning module employs topological mapping to encode intersection geometries as path nodes, thereby guiding the SSR pass through narrow shelf aisles. In real supermarkets, we evaluated the target tracking approach under shelf occlusions and the human following task within narrow aisles. Experimental results demonstrate that the visual-WEM tracking approach achieves a pose tracking accuracy of (4.56 mm, 2.98<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$^{\circ }$</tex-math></inline-formula>) under shelf occlusions. This study establishes the feasibility of human-robot collaboration in facilitating a hands-free shopping experience, highlighting its potential as a substitute for conventional shopping carts.

Digital microfluidics (DMF) is a versatile platform that enables parallel droplet operations and on-site programmable control, where droplet routing is a critical challenge for achieving automation. To enable automatic path generation from source to target electrodes and process-level feedback correction, this study proposes an adaptive intelligent control framework based on large language models (LLMs). The framework integrates language-guided topological map reconstruction, LLM-based path planning, and a dual closed-loop feedback mechanism to achieve high-precision droplet manipulation. To alleviate the spatial hallucination issues of LLMs in spatial reasoning such as discontinuous path planning and out-of-bound errors, a linguistic topological map reconstruction method is designed, improving path generation accuracy by 23.4% in complex obstacle environments compared to conventional spatial representations. In addition, a dual closed-loop feedback strategy is proposed, consisting of model self-correction feedback and camera-based image feedback, which enables real-time correction of routing deviations and enhances operational stability. Experimental results demonstrate that the proposed framework achieves 100% task success across various droplet distribution scenarios, significantly advancing the intelligence of microfluidic operations and providing a promising foundation for fully autonomous DMF systems.

Operating in inherently dynamic and unstructured environments, mobile robots must routinely navigate substantial uncertainties—particularly in crowded spaces or multi-door configurations. Yet mainstream navigation approaches often disregard such environmental unpredictability and underutilize historical task experience, leading to recurrent obstacle blockages that degrade both efficiency and safety. This work proposes an adaptive navigation method for uncertain environments using probabilistic topological mapping (PTM). We develop a Gaussian mixture model (GMM)-based PTM construction approach that extracts environmental feature nodes through path similarity and node density filtering, clusters nodes via GMM, and establishes topological connections to efficiently encode environmental characteristics. For adaptive navigation, we introduce: 1) a posterior heuristic algorithm for planning under uncertainty; 2) a topological edge analysis with re-planning and escape mechanisms for robustness; and 3) a dynamic particle filter for online probability updates. This framework enables robots to autonomously analyze environmental information and progressively optimize navigation strategies through probabilistic self-updating. Simulations and real-world experimental studies reveal that our method exhibits higher efficiency and safety compared to current mainstream navigation approaches.

ABSTRACT Dynamical systems have become a significant research area in recent years due to their rich structure and wide‐ranging applications. Analyzing their behavior often remains challenging, as even simple polynomial maps, such as the logistic map, can display highly complex dynamics. To circumvent this challenge, piecewise linear maps that are topologically conjugate to such polynomials are frequently considered. This article focuses on analyzing the dynamical behavior of a specific cubic polynomial. For this purpose, we first define a so‐called double‐sided tent map on a closed interval and represent it in base 2 to formulate its periodic points and investigate its dynamical properties. Furthermore, we construct a topological conjugacy map between the double‐sided tent map and the cubic polynomial, thereby demonstrating that the cubic polynomial also exhibits the same chaotic behavior as double‐sided tent map. We additionally prove that these dynamical systems possess periodic points of all even periods, but no periodic points of odd period except for a fixed point.

The rising energy demand of the building sector, coupled with the need for high indoor environmental quality (IEQ), necessitates intelligent control strategies. However, traditional methods like rule-based control (RBC) lack adaptability, while model predictive control (MPC) suffers from model dependency. Hence, this study designed and implemented an intelligent indoor environmental control system based on multi-sensor fusion and Building Information Modeling (BIM). At the algorithmic level, this study employed a deep fusion actor-critic reinforcement learning (DACRL) algorithm to achieve energy-efficient dynamic environmental control. The system employs a four-layer architecture with cloud-edge-end collaboration. Layer 1: The perception layer utilizes a multi-modal sensor network built on an STM32L0 low-power micro-controller and an SX1276 LoRa communication module. This architecture enables distributed collection and pre-processing of parameters such as temperature, humidity, CO2, light, and occupancy. Layer 2: The fusion layer proposes a spatio-temporal synchronization mechanism for multi-source asynchronous data based on sliding window dynamic weighting and an extended Kalman filter (EKF). As an optimization, an isolation forest and one-class support vector machine (SVM) are combined to construct a multi-layer anomaly detection process. Layer 3: The decision layer establishes a deep reinforcement learning control model based on the topology of BIM semantic mapping. The key novelty of this model lies in its innovative embedding of the Lieb–Thirring (L–T) inequality, which describes the energy concentration characteristics of spatial fields, as a spectral constraint into both the reward function and the online action projection process of the Actor-Critic algorithm. This model innovatively embeds the Lieb–Thirring (L–T) inequality, which describes the energy concentration characteristics of spatial fields, as a spectral constraint into the reward function and online action projection process of the Actor-Critic algorithm. Layer 4: The execution layer implements real-time closed-loop control of terminal equipment such as HVAC, lighting, and shading through the standard BMS protocol. This study designed large-scale simulation experiments covering six core scenarios: offices, classrooms, residences, hospitals, shopping malls, and data centers. The experimental results demonstrate that the system’s annual energy consumption per unit area decreases by up to 63.2% compared to traditional rule-based control (RBC) strategies and by 23.5% compared to model predictive control (MPC) strategies. At the same time, several indoor comfort metrics improved significantly, with CO2 concentrations falling below 800 ppm 91.5% of the time and the thermal comfort (PMV) index remaining within ± 0.3 94.2% of the time. The proposed DACRL algorithm demonstrates excellent convergence and generalization capabilities. It achieves stable convergence numerically with an average of only 15,000 training steps. Performance loss in cross-building type transfer testing is only 4.9%.

Infrastructure systems are essential for achieving the Sustainable Development Goals (SDGs), yet they face increasing risks from natural hazards—especially in low- and middle-income countries (LMICs), where infrastructure systems are often inadequate and institutional capacities are limited. This study presents a large-scale bibliometric analysis of 23,763 publications indexed in the Web of Science from 1960 to 2025, focused on the intersection of infrastructure and natural hazards in LMICs. We examined publication trends, author and study site geography, and thematic coverage of infrastructure and natural hazard types, while introducing sentiment analysis of publication metadata and topological mapping of research categories as novel approaches to capture research tone, urgency, and cross-domain linkages. Our findings reveal that although related publications have increased over time, LMICs remain significantly underrepresented in the literature, with only a few—such as Pakistan, Bangladesh, and Vietnam—receiving relatively more attention. This underrepresentation is mirrored in authorship patterns: participation from LMIC-based researchers is limited, and international collaboration is dominated by non-LMICs, with intra-LMIC partnerships particularly rare. Infrastructure topics such as social services, transportation, and water systems dominate the literature, while telecommunication, coastal protection, and green infrastructure are critically overlooked. Sentiment analysis suggests increasing concerns in hazard-prone regions such as Sub-Saharan Africa, North Africa, and the Middle East. This study highlights the need for more geographically inclusive, thematically balanced, and interdisciplinary research agendas. Expanding representation across LMICs, increasing local scholarly participation, addressing neglected infrastructure sectors, and recognizing region-specific challenges are essential for supporting more resilient and equitable infrastructure systems under intensifying natural hazard risks.