Abstract

Abstract
Highlights What are the main findings? PDAM improves hyperspectral image classification robustness under noisy labels. Prototype-guided dynamic masking and reconstruction outperform competing methods across multiple noise levels and datasets. What are the implications of the main findings? Prototype-based reliability estimation is effective for identifying trustworthy supervision in noisy hyperspectral data. Reliability-aware masking offers a practical strategy for robust remote-sensing learning with corrupted annotations.Highlights What are the main findings? PDAM improves hyperspectral image classification robustness under noisy labels. Prototype-guided dynamic masking and reconstruction outperform competing methods across multiple noise levels and datasets. What are the implications of the main findings? Prototype-based reliability estimation is effective for identifying trustworthy supervision in noisy hyperspectral data. Reliability-aware masking offers a practical strategy for robust remote-sensing learning with corrupted annotations.Abstract Existing noisy-label hyperspectral image classification (HSIC) methods usually address clean sample selection and representation regularization as separate problems, although the reliability of observed labels varies substantially across samples in hyperspectral data. This issue is amplified by mixed pixels, boundary ambiguity, spectral overlap, and limited labeled samples, which make hard clean samples difficult to distinguish from mislabeled ones. We therefore propose PDAM, a sample-reliability-guided training framework for noisy-label HSIC. The method first estimates feature-space class consistency by comparing each sample with the prototype of its observed class and converting this consistency into a reliability probability with a Gaussian mixture model. To reduce conservative false negatives, matched high-confidence selection is further used to recover hard but correctly labeled samples. The resulting reliability estimate then determines how strongly the observed label is trusted through target refinement and how strongly the input is perturbed through reliability-guided masking. Finally, masked reconstruction provides label-independent structural regularization so that uncertain samples can still contribute to spectral-spatial representation learning. Under the evaluated synthetic symmetric noise settings on the University of Pavia (UP), Salinas Valley (SV), and Kennedy Space Center (KSC) datasets, PDAM achieves the best OA and Kappa in most reported comparisons and improves robustness under both moderate and severe noise. At 30% noise, PDAM reaches 97.30% OA on UP, 98.13% OA on SV, and 95.37% OA on KSC. Ablation studies further support the necessity of reliability estimation, hard clean sample recovery, and reliability-guided supervision and regularization within this unified training mechanism.

Abstract

Abstract
Electric vehicles (EVs) are expected to accelerate the decarbonization of transport while also becoming a highly distributed and flexible resource for power systems. By coupling substantial battery storage with long parking times, EVs can support higher shares of renewable generation through controlled charging and, where available, bidirectional operation (e.g., V1G/V2G and related concepts). At the same time, large-scale EV uptake can increase peak demand, aggravate congestion and losses, and trigger voltage issues (particularly if charging remains unmanaged) potentially leading to costly network reinforcements. This chapter reviews the main EV types, charging modes and technologies (including fast and emerging wireless solutions), and the underlying storage technologies. It then discusses grid-integration architectures and operational strategies, from uncontrolled charging and time-of-use incentives to coordinated “smart charging” and V2G, highlighting their impacts on distribution networks and the requirements for communication, aggregation and system operator interaction. Finally, it outlines a future vision where EV flexibility is integrated with other distributed energy resources to provide local voltage support (active and reactive power), congestion management and frequency regulation services, enabled by appropriate standards, market mechanisms, and regulatory frameworks. © Springer Nature Switzerland AG 2026.

Abstract
As cities grow and sustainability becomes a key driver of urban policy, active modes of transport such as walking and cycling are increasingly promoted. However, current route planning applications rarely consider factors beyond time and distance. This paper presents the design and evaluation of a mobile application prototype that supports multicriteria route planning for active transport modes. The proposed solution incorporates user-defined weights for dimensions such as safety, comfort, accessibility, and environmental quality. To ensure adaptability and up-to-date information, the study also explores the feasibility of crowdsourcing as a complementary data source. A mixed-method approach was followed, including literature review, user surveys (n=242), interface prototyping, and usability testing with real users. The results demonstrate strong user interest in contributing to data updates, especially when motivated by non-monetary incentives such as gamified rankings. The final prototype was positively evaluated for usability and interface quality. This research confirms the potential of user-centered, crowdsourcing-enhanced route planning to improve the experience of active mobility users and support sustainable urban mobility goals. Copyright © 2025. Published by Elsevier B.V.

Abstract
Urban mobility in large cities faces increasing pressure due to population growth and congestion. Automated Fare Collection (AFC) systems offer a rich source of data for understanding public transport usage and informing data-driven improvements. This paper presents a case study in Fortaleza, Brazil, where we explore AFC smart card data to predict users’ next bus trips and travel volume. We develop a machine learning pipeline combining feature engineering and classification/regression models. A comparative evaluation of algorithms, including Random Forest, XGBoost, and Support Vector Machines, shows that decision-tree-based models achieve the best performance, particularly in handling noisy and imbalanced data. Our approach considers both user-level predictions and cluster-based analyses to improve model generalizability across user types. The results demonstrate the potential of AFC data to enhance transit planning, reduce overcrowding, and personalize mobility services. This study contributes to the growing body of research on smart mobility analytics in developing urban contexts. Copyright © 2025. Published by Elsevier B.V.