10784 research outputs found
Sort by
A Fail-Safe Decision Architecture for CCAM Applications
Full text linkPublisher Copyright: © The Author(s) 2026.In the context of Connected, Cooperative, and Automated Mobility (CCAM), precise ego-vehicle positioning and environmental status assessment are crucial. However, these tasks can be susceptible to sensor failures, misuse, and cyberattacks. Automation disengagements and system redundancy are common strategies to achieve Minimum Risk Conditions when failures occur. This paper presents a Fail-Safe decision architecture formulated within the framework of the SELFY project (https://selfy-project.eu/). The main aim is to reduce inaccuracies in GNSS-derived positioning through the incorporation of sensor fusion, AI-guided situational assessment, trajectory planning, and mode decision components. Additionally, the architecture has been designed to enable real-time updates and communication with external entities, including the Vehicle Security Operations Centre.Peer reviewe
SHINE-Fleet: Spanish Initiative for Sustainable Freight Transport Taking Advantage from Hydrogen and Automated Driving
Full text linkPublisher Copyright: © The Author(s) 2026.In recent years, reducing emissions in Freight-Transport has become a top priority in mobility. This paper presents the methodology for the full control loop of a truck’s powertrain, which has been transformed from a polluting GLP powered ICE to a zero-emission Hydrogen-powered electric powertrain within the SHINE-Fleet project. The adaptation has provided the ECU with increased flexibility to control the dynamics, enabling automated-driven missions like docking. The paper covers infrastructure-based localization for the vehicle, path planning to reach the setpoint, and control implementation with a Model Predictive Controller (MPC).Peer reviewe
A design framework for operationalizing trustworthy artificial intelligence in healthcare: Requirements, tradeoffs and challenges for its clinical adoption
Full text linkPublisher Copyright: © 2025 The Author(s).Artificial Intelligence (AI) holds great promise for transforming healthcare, particularly in disease diagnosis, prognosis, and patient care. The increasing availability of digital medical data, such as images, omics data, biosignals, and electronic health records, combined with advances in computing, has enabled AI models to approach expert-level performance. However, widespread clinical adoption remains limited, primarily due to challenges beyond technical performance, including ethical concerns, regulatory barriers, and lack of trust. To address these issues, medical AI systems must align with the principles of Trustworthy AI (TAI), which emphasize human agency and oversight, algorithmic robustness, privacy and data governance, transparency, bias and discrimination avoidance, and accountability. Yet, the complexity of healthcare processes (e.g., screening, diagnosis, prognosis, and treatment) and the diversity of stakeholders (clinicians, patients, providers, regulators) complicate the integration of TAI principles. To bridge the gap between TAI theory and practical implementation, this paper proposes a design framework to support developers in embedding TAI principles into medical AI systems. Thus, for each stakeholder identified across various healthcare processes, we propose a disease-agnostic collection of requirements that medical AI systems should incorporate to adhere to the principles of TAI. Additionally, we examine the challenges and tradeoffs that may arise when applying these principles in practice. To illustrate the discussion, we focus on cardiovascular diseases, which is a field marked by both high prevalence and active AI innovation, and demonstrate how TAI principles have been applied and where key obstacles persist.Peer reviewe
Multi-Rater Calibration Error Estimation
No full textPublisher Copyright: © The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.Calibration, the property of producing predicted probabilities that reflect true likelihoods of outcomes, is a relevant attribute of medical image computing models and a key requirement in clinical decision-making. However, empirical Calibration Error (CE) estimates suffer from instability in data-scarce scenarios. Here, for any existing CE we propose a Multi-Rater version of it (MR-CE), a wrapper over conventional calibration metrics, which provides a new strategy for estimating a CE that effectively addresses this limitation in situations where there are multiple annotations per sample. MR-CEs offer more consistent estimates of calibration errors by leveraging the consensus and disagreement among multiple annotators to generate virtually extended test datasets, more robust to typical binning artifacts. We evaluate a MR version of the popular Expected Calibration Error (ECE), and also of the more recent Kernel Density Estimation-ECE (kdeECE), in a comprehensive set of classification and segmentation problems, demonstrating improved stability compared to their single-rater CE counterparts. Specifically, we show that MR-CEs achieve a reduced variability as the test set size decreases across all analysed datasets. Our findings emphasize the critical role of modelling inter-rater variability not only for training but also for evaluating medical image analysis models, in particular when studying the calibration of modern neural networks.Peer reviewe
New Roads Governance Framework Based on Circularity
Full text linkPublisher Copyright: © The Author(s) 2026.One of the activities that can most influence climate neutrality is the activity of infrastructure construction. Some countries already include environmental concepts in the development of their infrastructures, as an approximation to neutrality, but this is insufficient, it is necessary to be more ambitious, and define in a broad and global way, a governance framework that includes this concept in detail, throughout their life cycle. To this end, a methodology is proposed that defines the neutrality of infrastructures based on circularity, including this concept in governance processes, through circularity indicators. In addition, to show its validity, it is applied to the contracting in the construction phase of the A-67 Highway project, Capacity Expansion, in Spain. Obtaining very good improvements in infrastructure governance, thus contributing to climate neutrality.Peer reviewe
Comparative Evaluation of Reinforcement Learning and Model Predictive Control for 6DoF Position Control of an Autonomous Underwater Vehicle
No full textPublisher Copyright: © The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.Autonomous Underwater Vehicles (AUVs) require precise and robust control strategies for 3D pose regulation in dynamic underwater environments. In this study, we present a comparative evaluation of model-free and model-based control methods for AUV position control. Specifically, we analyze the performance of neural network controllers trained by three Reinforcement Learning (RL) algorithms—Proximal Policy Optimization (PPO), Twin Delayed Deep Deterministic Policy Gradient (TD3), and Soft Actor-Critic (SAC)—alongside a Model Predictive Control (MPC) baseline. We train our RL methods in a simplified AUV simulator implemented in PyTorch, while our evaluation is done in a realistic marine robotics simulator called Stonefish. Controllers are evaluated on the basis of tracking accuracy, robustness to disturbances, and generalization capabilities. Our results show that, MPC suffers from unmodeled dynamics such as disturbances, whereas RL demonstrates adaptation capabilities to disturbances. Also, although MPC demonstrates strong control performance, it requires an accurate model, high compute power and a careful implementation to run in real-time whereas the control frequency of RL policies is only bound by the inference time of the policy network. Among RL-based controllers, PPO achieves the best overall performance, both in terms of training stability and control accuracy. This study provides insight into the feasibility of RL-based controllers for AUV position control, offering guidance for selecting suitable control strategies in real-world marine robotics applications.Peer reviewe
Hydrogen trapping and desorption spectra analysis in 300M ultra high strength martensitic steel - an experimental and modeling study
Full text linkPublisher Copyright: © 2025In the current work, the hydrogen diffusion and trapping in 300M steel were studied using the Kelvin probe and thermal desorption technique, respectively. Lattice diffusivity and activation energy for diffusion were obtained using two step permeation measurement at different temperatures. The activation energy for lattice diffusion in the material is 32kJ/mol, and the traps in the material are weak and reversible in nature with lower desorption energies (<20kJ/mol). The data obtained were used to model the diffusion and trapping behavior of hydrogen in the material. By combining continuum mechanics with finite element modeling, and integrating detailed deconvolution of thermal desorption spectra through a multi-trap diffusion framework, a rigorous methodology for the individualized optimization of detrapping parameters associated with each trap site in a complex multi-trap system is proposed. The optimized detrapping parameters were subsequently validated against experimental thermal desorption data across a range of heating rates.Peer reviewe
Benchmarking Machine Learning Models for QoE Estimation in Video Streaming: Accuracy, Efficiency, Confidence and Explainability
No full textPublisher Copyright: © The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.The accurate prediction of Quality of Experience (QoE) in video streaming services is essential for optimizing user satisfaction and network performance. While traditional Quality of Service (QoS) metrics provide objective measurements of network behavior, they often fail to reflect the subjective nature of user experience. This paper investigates the use of Machine Learning models to estimate QoE based on QoS indicators. Building upon the recently published SNESet dataset, we evaluate a range of modern regression techniques, including randomization-based neural networks, symbolic regression and Kolmogorov-Arnold Networks, alongside other traditional and ensemble-based models. A central focus of this study is the explainability of such new models, which enables the extraction of domain-relevant insights from the learned relationships. Using model-agnostic techniques for explainable Artificial Intelligence and uncertainty quantification, we assess the confidence of such models in their predictions and analyze the contribution of individual features to the estimated QoE. Our results underscore the need for explainable QoE prediction systems, closing the gap between data-driven modeling and domain expertise.Peer reviewe
Towards the Analysis of Software Supply Chain and EU Regulations
No full textPublisher Copyright: © The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.Software supply chain is becoming a relevant topic in cybersecurity, especially the software bill of materials (SBOM) in order to manage libraries and components dependencies. In addition, several European Union (EU) regulations have been approved in the context of cybersecurity. They provide horizontal cybersecurity requirements such as the Cyber Resilience Act (CRA). However, the link between SBOM and the EU regulations is not clear. Therefore, this paper provides an overview of the current literature’ state of the art in SBOMs and highlights its relationships with EU regulations. In fact, there is an evident increase of published research papers since the US executive order for improving Nation’s Cyber Security under the Biden’s administration, but there is scarce reference to legislations. Finally, we analyze the occurrence of key search strings within EU legislations.Peer reviewe
Reinforcement Learning in action: Powering intelligent intrusion responses to advanced cyber threats in realistic scenarios
Full text linkPublisher Copyright: © 2025 The AuthorsGiven the increasing incidence of sophisticated cyber-attacks, particularly Advanced Persistent Threats (APTs), there is a growing need for intelligent and adaptive intrusion response solutions. In this paper, we propose a Reinforcement Learning (RL)-based model for APT intrusion response that can manage dynamic, multi-stage attacks and large observation spaces. The model supports both policy-based and value-based learning approaches, enabling comparative evaluation between different strategies. We introduce a realistic RL training environment based on emulation infrastructure, which accurately reproduces APT scenarios using real systems and executes a wide range of authentic Intrusion Response System (IRS) actions. This setup includes time and variability constraints commonly encountered in operational environments, offering a more practical alternative to traditional simulations. The RL agents, implemented using Proximal Policy Optimization (PPO) and Deep Q-Network (DQN) algorithms, were both trained and evaluated within this industrial-style emulated environment. Empirical results demonstrate that both DRL algorithms successfully learned effective and well-timed defensive actions under realistic constraints, confirming their capability to operate in dynamic, real-world APT scenarios.Peer reviewe