HET sistemų centras / HET Systems Centre
Automated regulatory compliance for AI systems in the security domain: The case of dual-use deploymentItem type:Publikacija, research article[2026][S1b][T007,S001,S003][40]; ;Paskauskas, Richard; ; Open research Europe., 2026, p. 1-40The rapid expansion of the European Union’s digital regulatory framework and recent adoption of the Artificial Intelligence Act (AI Act) has introduced complex, overlapping compliance obligations for AI systems, and consequently, their developers and users. These challenges are amplified for dual-use AI systems, which may be developed for civilian markets yet deployed in military contexts, requiring alignment with both EU regulatory instruments and defence-specific frameworks such as NATO policies. Existing approaches to regulatory compliance remain largely manual, fragmented, and difficult to scale, particularly when legal requirements must be translated into actionable, system-level specifications for dual-use contexts. This paper proposes a novel ontology-based methodology for automated regulatory compliance requirements specification for dual-use AI systems. The methodology systematically integrates legal and technical perspectives by structuring compliance obligations across deployment domains (civilian and defence), system lifecycle phases, and requirement categories, including privacy- and security-related obligations. Implemented as a machine-readable knowledge graph using RDF/Turtle, the approach enables executable compliance modelling, where regulatory obligations are formalised as machine-interpretable entities that can be queried, validated, and deployment-specifically configured through semantic reasoning and SPARQL-based analysis. The methodology is validated through a detailed case study of an AI-powered espionage detection system, demonstrating how context-aware semantic reasoning and SHACL-based validation can ensure that regulatory requirements are consistently specified and mapped to concrete system components. The proposed framework advances the state of the art by providing a rigorous, extensible foundation for compliance-by-design and automated analysis, thereby reducing compliance risk and supporting responsible AI engineering in complex omni-use regulatory environments.
12Scopus© citavimai 2 Human-Centric Impacts of Cyberattacks on Autonomous Vehicle User Interactions: A Simulation-Based StudyItem type:Publikacija, research article[2026][P1g][T007][10]; ;Alam, Shahzad ;Bryans, JeremyJadidbonab, HesamaldinHuman Interaction and Emerging Technologies (IHIET-AI 2026): Artificial Intelligence and Future Applications., p. 497-506Connected and Autonomous Vehicles (CAVs) are rapidly advancing toward SAE Levels 4 and 5, fundamentally reshaping human–vehicle interaction. As these systems become increasingly automated and connected, cybersecurity incidents introduce risks that extend beyond technical failure, affecting user trust, perceived safety, and acceptance. This study investigates how passengers respond when cyberattacks occur during an autonomous ride-hailing service. A driving simulator study was conducted with 50 participants, replicating a robo-taxi experience along a 5-mile urban route. Participants experienced three scenarios: a control scenario, a passive attack scenario and an active attack scenario. Following each scenario, participants evaluated their trust in the system, perceived safety, perception of risk, privacy assurance, and service intention (intention to use). The results demonstrated statistically significant differences across the three scenarios. Passive attacks were associated with reduced trust and increased uncertainty, whereas active attacks produced stronger negative responses, including higher perception of risk and lower perceived safety. Participants also expressed concern about the reliability of the automated vehicle and their personal safety when abnormal behaviour occurred. Overall, the findings underscore that cyberattacks meaningfully influence how passengers evaluate the safety, reliability, and future adoption of autonomous mobility services. These results highlight the importance of integrating human-centric impact evaluation within cybersecurity risk assessment framework and the design of secure and trustworthy autonomous transport systems.
22 A dynamic simulation framework for mobile ad hoc networks in search and rescue operationsItem type:Publikacija, research article[2026][S1][T007][8]; ;Ghanem, Manal ;Correia-Hopkins, Sara ;Jones, Jonathan-LeeMicallef, NicholasSimulation modelling practice and theory., 2026, p. 29-36Network simulation is a cost-effective, widely adopted method for testing and improving new protocols and solutions in Mobile Ad Hoc Networks (MANETs). Despite the prevalence of simulation-based studies, the absence of unified, extensible frameworks limits the modelling of end-to-end communication and the generation of reusable datasets necessary for research reproducibility. Our paper introduces a simulation framework for evaluating MANETs in Search and Rescue (SAR) scenarios, developed entirely in NS-3 to streamline both simulation and dataset generation. The framework incorporates mobility modelling, energy tracking, communication dynamics, and quality of service (QoS) logging within a modular architecture. It implements heterogeneous nodes with role-specific behaviours and supports discovery mechanisms such as line-of-sight, beacon broadcasting, and signal-based detection. Two simulation versions are implemented using different propagation loss models to assess performance under ideal and obstructed conditions. Routing protocols, including AODV, OLSR, and DSDV, are evaluated across varying scalability levels. Our results confirm established trends in MANET routing and demonstrate the framework’s flexibility in simulating mission-driven environments. The work presents multiple datasets and a reusable, extensible, scenario-driven simulation environment for researchers investigating MANET behaviour in complex applications such as disaster relief and tactical communications.
4Scopus© citavimai 1 Toward sustainable supply chains: integrating digital technologies for Scope 3 emission reduction and cyber-physical resilienceItem type:Publikacija, research article[2026][S1][S004,T007][15]Environment Systems and Decisions., 2026, p. 307-321Scope 3 greenhouse gas (GHG) emissions and cyber-physical resilience are typically governed through separate institutional and operational frameworks, despite their growing interdependence in digitally mediated supply chains. This separation creates an execution gap that fragments decarbonization, resilience, and investment decisions across data architectures and organizational functions. Grounded in resilience theory and a transdisciplinary systems perspective, this study advances the testable hypothesis that institutional separation of Scope 3 emissions management and cyber-physical resilience is associated with lower system performance, whereas integration through digital governance mechanisms improves operational resilience, emissions measurement accuracy, and financial decision coherence. System performance is operationalized through three dimensions: (1) supply-chain observability and data integration; (2) resilience capacity across the prepare–absorb–recover–adapt cycle; and (3) coherence of investment decision-making under dual materiality. The study applies a structured integrative review with analytical triangulation across sectoral datasets, cyber-risk distributions, and governance frameworks, complemented by an illustrative application in emergency medical services (EMS) in Lithuania. Findings indicate that Scope 3 burdens and cyber-physical risks co-locate in supplier-intensive and digitally dependent sectors, revealing systemic vulnerability; and that effective integration requires a joint governance architecture linking digital observability, standards, financial evaluation, and human-centered decision-making. The article contributes an integrative framework that reframes decarbonization and cyber-physical resilience as a unified systems problem and provides an operational basis for coordinated governance across sustainable supply chains.
2Scopus© citavimai 1 The role of digital transformation adaptation in linking self-determination theory to workforce motivation in construction 5.0Item type:Publikacija, research article[2026][S1][S003,T002][22]; Basit Khan, MuhammadEngineering, construction and architectural management, 2026, vol. 60, p. 1-22This study explores the direct and mediating relationships among Self-Determination Theory (SDT), Digital Transformation Adaptation (DTA 5.0) and Workforce Motivation Outcomes (WMO) in the context of Construction 5.0. The primary objective is to examine how SDT influences DTA 5.0 and, in turn, how DTA 5.0 impacts WMO.
Scopus© citavimai 2 Performance benchmarking of scan-to-HBIM workflows for cultural heritage (2012–2025)Item type:Publikacija, research article[2026][S1][T002][11]Automation in construction, 2026, vol. vol. 184, p. 350-360This paper conducts a comprehensive systematic review of Scan-to-HBIM workflows from 2012 to 2025, examining methodological structures, algorithmic support, and application maturity within digital workflows for cultural heritage documentation and management. The review synthesises global literature to classify contributions across technical domains: geometric acquisition, condition assessment, AI-assisted modelling, performance benchmarking, semantic enrichment, and immersive VR/AR applications, alongside the emerging cultural value dimension. The findings indicate a progressive shift toward hybrid human–algorithm workflows, combining deep learning, rule-based modelling, and semantic knowledge structures, rather than fully automated pipelines. Accuracy benchmarking highlights persistent inconsistencies in reporting frameworks across RMSE, IoU, and level-of-detail metrics, underscoring the need for greater methodological transparency. The paper concludes by identifying standardisation, interoperable data governance, and selective, scalable semantic AI integration as critical priorities for advancing reliable and sustainable Scan-to-HBIM practice in cultural heritage contexts.
Scopus© citavimai 3 Deep sequence learning with multi-task supervision for scalable population health monitoringItem type:Publikacija, research article[2026][S1a][N009][21] ;Ashraf, Imran; ;Awais, Muhammad ;Bouchelligua, Wided ;Mansour, Sahar ;Alyounis, EssaNawaz, MajedFrontiers in public health, 2026, vol. 14, p. 1-21Models designed for artificial intelligence-driven population health monitoring must be able to integrate multiple types of information over time, be heterogeneous in the data they consider, and be deployed at scale. This research proposes an integrated multitask framework for predicting population disease incidence and mortality risk, based on deep learning and using both longitudinal biobank and National Health Survey data in a scalable manner. The framework will learn shared temporal representations from clinical lab-generated/demographically defined survey variables by applying a strict prospective evaluation approach across the framework. On UK Biobank, the proposed model achieves an AUROC of 0.842 and a C-index of 0.791, outperforming strong baselines such as Proformer having AUROC 0.829 and UKB-MDRMF having AUROC 0.821. On NHANES, the model achieves an AUROC of 0.861 and a C-index of 0.808, exceeding those of commonly used boosted tree and survival-based approaches. Temporal robustness analysis shows controlled performance degradation across extended horizons, with UK Biobank AUROC decreasing from 0.872 at 1 year to 0.819 at 10 years and NHANES AUROC decreasing from 0.886 at 1 year to 0.836 at 5 years. Subgroup evaluation demonstrates stable performance across age, sex, and clinical risk strata, while scalability analysis confirms near-linear training growth and stable inference latency below three milliseconds per individual at biobank scale.
1 Addressing Social Innovation and Sustainability Transitions in Increasingly Complex and Co-Dependent Human, environmental, and Technological Systems: A Proposal for an Experimental Intervention in AcademiaItem type:Publikacija, conference paper[2024][T2][S003][2]; Fourth ERUA Annual Summit 2024 : Social Innovations for Transformative Society, 25th-27th June 2024, [Mykolas Romeris university] : Abstracts, Day 1 – 25 June 2024., p. 6-724