Search Results (819)

Unstructured POI name texts are widely used in fine-grained urban analysis, yet missing labels and semantic ambiguity often limit their value for spatial inference. This study proposes a large language model-based semantic–spatial inference framework (LLM-SSIF), a lightweight semantic–spatial pipeline that translates POI texts into interpretable, fine-grained spatial evidence through an end-to-end workflow that couples scalable label expansion with scale-controlled spatial diagnostics at a 500 m resolution. A key advantage of LLM-SSIF is its deployability: LoRA-based parameter-efficient fine-tuning of an open LLM enables lightweight adaptation under limited compute while scaling fine-label coverage. Trained on a nationwide cuisine-labeled dataset (~220,000 records), the model achieves strong multi-class short-text recognition (macro-F1 = 0.843) and, in the Guangzhou–Shenzhen demonstration, expands usable fine-category labels by ~14–15× to support grid-level inference under long-tail sparsity. The spatial module then isolates cuisine-specific over/under-representation beyond overall restaurant intensity, revealing contrasting cultural configurations between Guangzhou and Shenzhen. Overall, LLM-SSIF provides a reproducible and transferable way to translate unstructured POI texts into spatial–statistical evidence for comparative urban analysis. Full article

Intelligent Transportation Systems (ITS) have progressively incorporated machine learning to optimize traffic efficiency, enhance safety, and improve real-time decision-making. However, the traditional centralized machine learning (ML) paradigm faces critical limitations regarding data privacy, scalability, and single-point vulnerabilities. This study explores FL as a decentralized alternative that preserves privacy by training local models without transferring raw data. Based on a systematic literature review encompassing 39 ITS-related studies, this work classifies applications according to their architectural detail—distinguishing systems from models—and identifies three families of federated learning (FL) frameworks: privacy-focused, integrable, and advanced infrastructure. Three representative frameworks—Federated Learning-based Gated Recurrent Unit (FedGRU), Digital Twin + Hierarchical Federated Learning (DT + HFL), and Transfer Learning with Convolutional Neural Networks (TFL-CNN)—were comparatively analyzed against a client–server baseline to assess their suitability for ITS adaptation. Our qualitative, architecture-level comparison suggests that DT + HFL and TFL-CNN, characterized by hierarchical aggregation and edge-level coordination, are conceptually better aligned with scalability and stability requirements in vehicular and traffic deployments than pure client–server baselines. FedGRU, while conceptually relevant as a meta-framework for coordinating multiple organizational models, is primarily intended as a complementary reference rather than as a standalone architecture for large-scale ITS deployment. Through application-level evaluations—including traffic prediction, accident detection, transport-mode identification, and driver profiling—this study demonstrates that FL can be effectively integrated into ITS with moderate architectural adjustments. This work does not introduce new experimental results; instead, it provides a qualitative, architecture-level comparison and adaptation guideline to support the migration of ITS applications toward federated learning. Overall, the results establish a solid methodological foundation for migrating centralized ITS architectures toward federated, privacy-preserving intelligence, in alignment with the evolution of edge and 6G infrastructures. Full article

To address the scientific gap concerning optimal urban wind turbine morphology, this work presents an experimental performance comparison between two small-scale wind turbine designs: a conventional three-blade horizontal-axis wind turbine (HAWT) and a duct-equipped Ayanz-inspired screw-blade turbine. Both configurations were tested in a controlled wind tunnel under steady and transient wind conditions, including synthetic gusts designed to emulate urban wind patterns. The analysis focuses on power output, aerodynamic efficiency (via the power coefficient $C_P$), dynamic responsiveness, and integration suitability. A key novelty of this study lies in the full-scale experimental comparison between a non-conventional Ayanz screw-blade turbine and a standard three-blade turbine, since experimental data contrasting these two geometries under both steady and gusty urban wind conditions are extremely scarce in the literature. Results show that while the three-blade turbine achieves a higher $C_P$ peak and greater efficiency near its optimal operating point, the Ayanz turbine exhibits a broader performance plateau and better self-starting behavior under low and fluctuating wind conditions. The Ayanz model also demonstrated smoother power build-up and higher energy capture under specific gust scenarios, especially when wind speed offsets were low. Furthermore, a methodological contribution is made by comparing the $C_P$ vs. tip speed ratio λ curves at multiple wind speeds, providing a novel framework (plateau width analysis) for realistically assessing turbine adaptability and robustness to off-design conditions. These findings provide practical insights for selecting turbine types in variable or urban wind environments and contribute to the design of robust small wind energy systems for deployments in cities. Full article

Accurate detection and localization of traffic objects are essential for autonomous driving tasks such as path planning. While semantic segmentation is able to provide pixel-level classification, existing networks often fail under challenging conditions like nighttime or rain. In this paper, we introduce a new training framework that combines unsupervised domain adaptation with high dynamic range imaging. The proposed network uses labeled daytime images along with unlabeled nighttime HDR images. By utilizing the fine details typically lost in conventional SDR images due to dynamic range compression, and incorporating the UDA training strategy, the framework effectively trains a model that is capable of semantic segmentation across adverse weather conditions. Experiments conducted on four datasets have demonstrated substantial improvements in inference performance under nighttime and rainy scenarios. The accuracy for daytime images is also enhanced through expanded training diversity. Full article

Autonomous lane-changing is one of the most critical and complex tasks in automated driving. Recent progress in reinforcement learning (RL) has shown strong potential to help autonomous vehicles (AVs) make safe and flexible lane-change decisions in real time under uncertain traffic conditions. In the current studies, there is a lack of a common structure that links RL algorithms, simulation tools, and performance evaluation methods. This paper presents a detailed examination of RL-based lane-changing systems in AVs, tracing their development from early rule-based models to modern learning-based approaches. It introduces a clear classification of lane-changing types—discretionary, mandatory, cooperative, and emergency—and connects each to the most suitable RL methods, including value-based, policy-based, actor–critic, model-based, and hybrid algorithms. Each method is examined for its performance, safety, and computational demands. Furthermore, it reviews major simulation environments, such as SUMO, CARLA, and SMARTS, and summarizes key evaluation measures related to safety, efficiency, comfort, and real-time performance. The comparison shows open research challenges, including model adaptation, safety assurance, and transfer from simulation to real-world driving. Finally, it outlines promising directions for future work, such as cooperative decision-making, safe and explainable RL, and lightweight models for real-time use. This review provides a clear foundation and practical guide for developing reliable and understandable RL-based lane-changing systems for future intelligent transportation. Full article

Floods are among the most destructive natural disasters, posing major challenges to human safety, property, and urban resilience. Effective communication of flood risk is therefore crucial for disaster preparedness and the sustainable management of smart cities. This study explores how interface design elements of flood hazard maps, including interaction modes and legend color schemes, influence users’ risk perception, decision support, and usability. An online questionnaire survey (N = 776) and a controlled 2 × 2 experiment (N = 40) were conducted to assess user comprehension, cognitive load, and behavioral responses when interacting with different visualization formats. Results show that slider-based interaction significantly reduces task completion and map-reading times compared with drop-down menus, enhancing usability and information efficiency. Multicolor legends, although requiring higher cognitive effort, improve users’ risk perception, engagement, and memory of flood-related information. These findings suggest that integrating cognitive principles into interactive design can enhance the effectiveness of digital disaster communication tools. By combining human–computer interaction, visual cognition, and smart governance, this study provides evidence-based design strategies for developing intelligent and user-centered flood hazard mapping systems. The proposed framework contributes to the advancement of smart urban resilience and supports the broader goal of building safer and more sustainable cities. Full article

Air pollution is a growing environmental issue in densely populated urban areas worldwide. Rapid population growth and the consequent increase in energy demand, emissions from industrial activities and vehicular traffic, and the reduction in vegetation cover have in recent years led to increasing concerns about quality of life, especially due to serious health problems associated with respiratory diseases. This study focuses on air quality in the city of Turin in north-western Italy. Continuous one-year monitoring, which collected approximately two million georeferenced data points, was possible using specific devices—palm-sized, wearable, and commercially available sensors—in different parts of the city. This enabled the assessment of the geographical and seasonal distributions of the most commonly studied air pollutants, namely particulate matter (PM) of three size fractions, nitrogen dioxide (NO₂), and total volatile organic compounds (TVOCs). The results highlight that the north-western zone and the urban centre are the most polluted areas. In particular, seasonal variations suggest that space heating and cooling systems, together with industrial activities, are the main contributors, more so than vehicular traffic. In this context, handheld devices in an IoT context can provide a reliable description of the spatial and temporal distribution of common air pollutants. Full article

Metropolitan service-placement optimization is computationally challenging under strict evaluation budgets and regulatory constraints. Existing approaches either neglect capacity constraints, producing infeasible solutions, or employ population-based metaheuristics requiring hundreds of evaluations—beyond typical municipal planning resources. We introduce a two-stage optimization framework combining Bayesian optimization with domain-informed heuristics to address this constrained, mixed discrete–continuous problem. Stage 1 optimizes continuous service area allocations via the Tree-structured Parzen Estimator with empirical gradient prioritization, reducing effective dimensionality from 81 services to 10–15 per iteration. Stage 2 converts allocations into discrete unit placements via efficiency-ranked bin packing, ensuring regulatory compliance. Evaluation across 35 benchmarks on Saint Petersburg, Russia (117–3060 decision variables), demonstrates that our method achieves 99.4% of the global optimum under a 50-evaluation budget, outperforming BIPOP-CMA-ES (98.4%), PURE-TPE (97.1%), and NSGA-II (96.5%). Optimized configurations improve equity (Gini coefficient of 0.318 → 0.241) while maintaining computational feasibility (2.7 h for 109-block districts). Open-source implementation supports reproducibility and facilitates adoption in metropolitan planning practice. Full article

Cloud-based platforms form the backbone of smart city ecosystems, powering essential services such as transportation, energy management, and public safety. However, their operational complexity generates vast volumes of system logs, making manual anomaly detection infeasible and raising reliability concerns. This study addresses the challenge of data scarcity in log anomaly detection by leveraging Large Language Models (LLMs) to enhance domain-specific classification tasks. We empirically validate that domain-adapted classifiers preserve strong natural language understanding, and introduce a Proximal Policy Optimization (PPO)-based approach to align semantic patterns between LLM outputs and classifier preferences. Experiments were conducted using three Transformer-based baselines under few-shot conditions across four public datasets. Results indicate that integrating natural language analyses improves anomaly detection F1-Scores by 5–86% over the baselines, while iterative PPO refinement boosts classifier’s “confidence” in label prediction. This research pioneers a novel framework for few-shot log anomaly detection, establishing an innovative paradigm in resource-constrained diagnostic systems in smart city infrastructures. Full article

On-demand public transport systems are increasingly adopted to improve service flexibility, reduce operating costs, and meet emerging mobility needs. Evaluating their performance under realistic demand and operational conditions, however, remains a complex challenge. This study presents a hybrid simulation framework that integrates deep learning-based demand forecasting, behavioural survey data, and agent-based simulation to assess system performance. A BiLSTM neural network trained on real-world smartcard data forecasts short-term passenger demand, which is embedded into an agent-based model simulating vehicle dispatch, routing, and passenger interactions. The framework is applied to a case study in Melbourne, Australia, comparing a baseline fixed-route service with two on-demand scenarios. Results show that the most flexible scenario reduces the average passenger trip time by 32%, decreases the average wait time by 34%, increases vehicle occupancy from 12.1 to 18.6 passengers per vehicle, lowers emissions per passenger trip by 72%, and cuts the service cost per trip from AUD 6.82 to AUD 4.73. These findings demonstrate the potential of hybrid on-demand services to improve operational efficiency, passenger experience, and environmental outcomes. The study presents a novel, integrated methodology for scenario-based evaluation of on-demand public transportation using real-world transportation data. Full article

Smart cities generate vast multimodal data from IoT devices, surveillance systems, health monitors, and environmental monitoring infrastructure. The seamless integration and interpretation of such multimodal data is essential for intelligent decision-making and adaptive urban services. Multimodal machine learning (MML) provides a unified framework to fuse and analyze diverse sources, surpassing conventional unimodal and rule-based approaches. This review surveys the role of MML in smart city sensing across mobility, public safety, healthcare, and environmental domains, outlining key data modalities, enabling technologies and state-of-the-art fusion architectures. We analyze major methodological and deployment challenges, including data alignment, scalability, modality-specific noise, infrastructure limitations, privacy, and ethics, and identify future directions toward scalable, interpretable, and responsible MML for urban systems. This survey serves as a reference for AI researchers, urban planners, and policymakers seeking to understand, design, and deploy multimodal learning solutions for intelligent urban sensing frameworks. Full article

Cities increasingly utilise digital technologies to tackle climate risks and urban shocks, yet their real impact on resilience remains uncertain. This paper systematically reviews 115 peer-reviewed studies (2012–2024) to explore how smart city technologies engage with planning instruments, governance arrangements, and social processes, following PRISMA 2020 and combining bibliometric co-occurrence mapping with a qualitative synthesis of full texts. Three themes organise the findings: (i) urban planning and design, (ii) smart technologies in resilience, and (iii) strategic planning and policy integration. Across these themes, Internet of Things (IoT) and geographic information system (GIS) applications have the strongest empirical support for enhancing absorptive and adaptive capacities through risk mapping, early warning systems, and infrastructure operations, while artificial intelligence, digital twins, and blockchain remain largely at pilot or conceptual stages. The review also highlights significant geographical and hazard biases: most cases come from high-income cities and concentrate on floods and earthquakes, while slow stresses (such as heat, housing insecurity, and inequality) and cities in the Global South are under-represented. Overall, the study promotes a “smart–resilience co-production” perspective, demonstrating that resilience improvements rely less on technology alone and more on how digital systems are integrated into governance and participatory practices. Full article

A sustainable city requires a sustainable means of transportation. This ambition is leading towards higher penetration of electric vehicles (EVs) in our cities, in both the private and commercial sectors, putting an ever greater burden on the existing power grid. Modern deregulated power grids vary electricity tariffs from location to location and from time to time to compensate for any additional burden. In this paper, we propose a profit-aware solution to strategically manage the movements of EVs in the city to support the grid while exploiting these locational, time-varying prices. This work is divided into three parts: (M1) profit-aware charging location and optimal route selection, (M2) profit-aware charging and discharging location and optimal route selection, and (M2b) profit-aware charging and discharging location and optimal route selection considering demand-side flexibility. This work is tested on the MATLAB programming platform using the Gurobi optimisation solver. From the extensive case studies, it is found that M1 can yield profits up to 2 times greater than those of its competitors, whereas M2 can achieve profits up to 2.5 times higher and simultaneously provide substantial grid support. Additionally, the M2b extension makes M2 more efficient in terms of grid support. Full article

The development of autonomous tunneling is crucial for building the intelligent underground infrastructure that smart cities require. However, in complex urban environments, the need for frequent manual intervention during Tunnel Boring Machine (TBM) operation remains a challenge, hindering overall efficiency and safety. To address the human–machine collaboration gap, this study analyzes practical experiences from six tunnel projects that use autonomous driving systems. Building on this foundation, we develop an intelligent assistant powered by a large language model (LLM). The assistant constructs a complete service architecture and intervention mechanism, proposes a phased intention recognition framework, and uses conversational interaction to achieve efficient human–machine communication. Experimental results demonstrate the strong classification performance of our intention recognition model. Furthermore, engineering case studies validate the assistant’s effectiveness in enhancing operational transparency, increasing user trust, bridging the human–machine information gap, and ultimately ensuring safer and more reliable tunneling. This research provides a feasible and innovative technological path for human–machine collaboration in the construction of critical urban infrastructure. Full article

Smart cities seek to improve urban living by embedding advanced technologies into infrastructures, services, and governance. Edge Artificial Intelligence (Edge AI) has emerged as a critical enabler by moving computation and learning closer to data sources, enabling real-time decision-making, improving privacy, and reducing reliance on centralized cloud infrastructure. This survey provides a comprehensive review of the foundations, challenges, and opportunities of edge AI in smart cities. In particular, we begin with an overview of layer-wise designs for edge AI-enabled smart cities, followed by an introduction to the core components of edge AI systems, including applications, sensing data, models, and infrastructure. Then, we summarize domain-specific applications spanning manufacturing, healthcare, transportation, buildings, and environments, highlighting both the softcore (e.g., AI algorithm design) and the hardcore (e.g., edge device selection) in heterogeneous applications. Next, we analyze the sources of sensing data generation, model design strategies, and hardware infrastructure that underpin edge AI deployment. Building on these, we finally identify several open challenges and provide future research directions in this domain. Our survey outlines a future research roadmap to advance edge AI technologies, thereby supporting the development of adaptive, harmonic, and sustainable smart cities. Full article