Periodic Reporting for period 2 - EURMARS (An advanced surveillance platform to improve the EURopean Multi Authority BordeR Security efficiency and cooperation)
Période du rapport: 2024-04-01 au 2025-09-30
Security risks and threats in the maritime domain are becoming increasingly more complex day by day. Within the EU there have been significant increases in irregular migration flows and human trafficking and smuggling but also of other illegal activities, such as drugs and arms trafficking and illegal, unreported and unregulated fishing. The European Union Maritime Security Strategy, and other EU initiatives such as EUROSUR and CISE, highlight that closer collaboration between authorities at national, regional and EU levels can constitute the key action to increase both situational awareness and operational efficiency. In particular, there is increased interest in assessing the viability of multi-authority cooperation in (especially high altitude) sensing and capability advancement for enhanced wide area surveillance of land, air and sea borders. EURMARS vision is to expand the common risk assessment practices currently deployed by authorities to enable the development, deployment and evaluation of a secure multitasking surveillance platform that improves sensing capabilities for a wide range of security risks and threats in wider border areas by clustering high altitude platforms technology, satellite imagery, UAVs and ground-based sensors into a novel joint surveillance capability. As part of the EURMARS framework, the various existing and future systems for maritime surveillance will be integrated, to allow for the collaborative operation and the provision of the sensing results to related authorities. The open architecture will build on the lessons learnt of previous initiatives, assimilate the knowledge of the stakeholders and their practice on CISE and other relevant systems, exploit the latest AI, risk assessment and visualization innovations, and undergo extensive technical and user acceptance tests and ethical and legal impact assessments.
Throughout EURMARS, the consortium delivered a coordinated programme resulting in a next-generation, multi-authority surveillance platform for European border security. The project began with a broad requirements-engineering phase involving authorities and technical partners, producing full technical, user, functional, non-functional and ethical requirements, along with the assessment methodology, KPIs and acceptance criteria. Building on these, the team designed and refined a modular architecture supporting heterogeneous sensors, advanced analytics, CISE/VDES interoperability, AI components and layered decision-support.
Major work focused on enhancing sensing capabilities. EURMARS developed and upgraded ground, low-altitude, high-altitude, UAV and satellite platforms, including fog-resilient SWIR systems, SMART SENSE mast solutions, multi-camera UAV payloads, AI-based detection/tracking networks, high-altitude 3D-LiDAR simulations and GEO-satellite vessel-tracking concepts. Real-time pipelines for data acquisition, processing, behaviour analysis and anomaly detection were implemented using embedded GPU devices and state-of-the-art deep learning. A robust common data hub and digital interfaces enabled seamless links with CISE, VDES, Copernicus and maritime databases, supported by multimodal fusion using fuzzy logic, neural inference and contextual meta-fusion.
The Command-and-Control and decision-support environment integrated all sensing streams into a unified, web-based, GIS-driven platform with real-time awareness sharing, secure communication, alarms and operator workflows. CIRAM was operationalised into an automated risk engine that prioritises threats and supports decision-making. System integration involved multi-phase validation, cyber-physical security assessments, stress testing and the creation of a comprehensive security framework.
Field activities formed the culmination of the work. Two prototype iterations were deployed in the Bulgarian living lab, generating a large multimodal dataset for training and benchmarking. Full demonstrations followed in Cyprus, the UK and the Bulgaria–Romania cross-border area. The platform achieved strong detection accuracy, solid performance in adverse maritime environments, seamless multi-sensor integration and effective cross-border cooperation, meeting or surpassing most KPIs.
Complementary efforts ensured ethical oversight, AI-Act alignment and continuous monitoring by ethics mentors. Dissemination and exploitation activities delivered communication materials, policy inputs and industrial engagement, while stakeholder involvement increased visibility and operational relevance. Overall, EURMARS delivered a mature, interoperable, AI-enabled surveillance ecosystem that strengthens situational awareness, operational coordination and border-security capabilities across sea, land and air domains.
Major work focused on enhancing sensing capabilities. EURMARS developed and upgraded ground, low-altitude, high-altitude, UAV and satellite platforms, including fog-resilient SWIR systems, SMART SENSE mast solutions, multi-camera UAV payloads, AI-based detection/tracking networks, high-altitude 3D-LiDAR simulations and GEO-satellite vessel-tracking concepts. Real-time pipelines for data acquisition, processing, behaviour analysis and anomaly detection were implemented using embedded GPU devices and state-of-the-art deep learning. A robust common data hub and digital interfaces enabled seamless links with CISE, VDES, Copernicus and maritime databases, supported by multimodal fusion using fuzzy logic, neural inference and contextual meta-fusion.
The Command-and-Control and decision-support environment integrated all sensing streams into a unified, web-based, GIS-driven platform with real-time awareness sharing, secure communication, alarms and operator workflows. CIRAM was operationalised into an automated risk engine that prioritises threats and supports decision-making. System integration involved multi-phase validation, cyber-physical security assessments, stress testing and the creation of a comprehensive security framework.
Field activities formed the culmination of the work. Two prototype iterations were deployed in the Bulgarian living lab, generating a large multimodal dataset for training and benchmarking. Full demonstrations followed in Cyprus, the UK and the Bulgaria–Romania cross-border area. The platform achieved strong detection accuracy, solid performance in adverse maritime environments, seamless multi-sensor integration and effective cross-border cooperation, meeting or surpassing most KPIs.
Complementary efforts ensured ethical oversight, AI-Act alignment and continuous monitoring by ethics mentors. Dissemination and exploitation activities delivered communication materials, policy inputs and industrial engagement, while stakeholder involvement increased visibility and operational relevance. Overall, EURMARS delivered a mature, interoperable, AI-enabled surveillance ecosystem that strengthens situational awareness, operational coordination and border-security capabilities across sea, land and air domains.
EURMARS introduced technological and operational innovations that significantly advance European border-surveillance capabilities beyond the state of the art. It delivered unprecedented multimodal situational awareness by integrating coastal sensors, UAVs, high-altitude platforms, GEO-based observation and Copernicus imagery into a single real-time operational environment. This combined surface, low-altitude, wide-area and orbital perspectives, an integration not previously implemented in border management. The project developed advanced AI-driven detection, tracking and behaviour-analysis models for embedded systems, enabling real-time recognition and behavioural interpretation under harsh maritime conditions, moving well beyond traditional optical- or radar-only systems.
The data-fusion framework represents a major advance. It synthesises heterogeneous sources while addressing asynchronous updates, noisy measurements, alignment errors and missing data. Adaptive inference mechanisms dynamically request needed sensor inputs, adjust fusion weights and use contextual heuristics to refine threat interpretation, producing a more resilient and accurate operational picture than single-modality solutions.
At the operational level, EURMARS transformed collaborative border management by converting the manual CIRAM framework into an automated, data-driven risk engine capable of ranking threats, forecasting escalation and guiding operators through structured workflows. The collaborative web-based C2 environment enables authorities to share insights, alerts and operational layers across borders in real time.
The project also advanced high-altitude and space-based surveillance through 3D-LiDAR simulations for Stratobus-like platforms and GEO-satellite concepts for persistent wide-area vessel tracking. These approaches expand coverage and endurance far beyond current airborne or satellite systems.
Finally, EURMARS set new standards for ethical and regulatory compliance of high-risk AI in border management. It produced an AI-Act-aligned compliance blueprint, practical guidelines for explainability and transparency, and privacy- and security-by-design architectures. These outcomes strengthen European leadership in trustworthy, responsible security-focused AI.
The data-fusion framework represents a major advance. It synthesises heterogeneous sources while addressing asynchronous updates, noisy measurements, alignment errors and missing data. Adaptive inference mechanisms dynamically request needed sensor inputs, adjust fusion weights and use contextual heuristics to refine threat interpretation, producing a more resilient and accurate operational picture than single-modality solutions.
At the operational level, EURMARS transformed collaborative border management by converting the manual CIRAM framework into an automated, data-driven risk engine capable of ranking threats, forecasting escalation and guiding operators through structured workflows. The collaborative web-based C2 environment enables authorities to share insights, alerts and operational layers across borders in real time.
The project also advanced high-altitude and space-based surveillance through 3D-LiDAR simulations for Stratobus-like platforms and GEO-satellite concepts for persistent wide-area vessel tracking. These approaches expand coverage and endurance far beyond current airborne or satellite systems.
Finally, EURMARS set new standards for ethical and regulatory compliance of high-risk AI in border management. It produced an AI-Act-aligned compliance blueprint, practical guidelines for explainability and transparency, and privacy- and security-by-design architectures. These outcomes strengthen European leadership in trustworthy, responsible security-focused AI.