AAAM2: Automated Arms and Ammunition Mapping

Context

The rapid detection, identification, documentation and mapping of explosive ordnance and illicit arms is critical for security in conflict zones, high-crime areas, and transnational enforcement contexts. Traditional methods relying on human analysts are slow, costly, and prone to error, limiting situational awareness and operational effectiveness. Advanced technologies, including computer vision and artificial intelligence (CV-AI), offer significant potential to improve accuracy, efficiency, and safety while enabling data-driven monitoring and mapping of arms and ammunition. Phase 1 of the Automated Arms and Ammunition Mapping project demonstrated the feasibility of developing CV-AI tools for practical deployment. These tools integrate high-performance algorithms into mobile phones, drones, and cameras to detect and identify unexploded ordnance, illicit weapons, and ammunition. By combining AI-driven analysis with geo-spatial reference data, the project enables granular mapping of arms and ammunition distribution patterns. Phase 1 also incorporated a strong digital evidence component to ensure compliance with national and international accountability standards.

Building on these results, Phase 2 will scale up detection capabilities to cover a wider range of weapons and expand deployment through partnerships with law enforcement, counter-trafficking, and counter-terrorism actors, creating synergies with existing EU counter-proliferation initiatives.

Overall objective

This project aims to make explosive ordnance (EO) detection, identification, recognition, mapping, and data collection more accurate, safer, and more effective through the application of new AI and computer vision technologies.

Specific objectives

• To scale-up of the number of deployable high-performance CV-AI detection and identification algorithms for explosive ordnance and firearm ammunition.
• To build mobile deployment tools for illicit arms and ammunition detection, identification and mapping.
• To ensure ongoing integration and interoperability with other EU-supported counter-trafficking, counter-proliferation and related initiatives.

Concrete activities

• Production of 100 high-performance explosive ordnance detectors for mobile devices and drones, with associated digital assets.
• Development and field-testing of a drone-based platform for the detection, identification, and mapping of explosive ordnance.
• Implementation and deployment of a smartphone app (iOS and Android) integrating detectors, hazard tagging, and geo-spatial capabilities for authorised actors.
• Establishment of a Beta version of the Small Calibre Ammunition Reference Table (SMART) and the ‘Ammo-ID’ mobile detection tool, populated with 1,000 unique ammunition profiles.
• Assessments of ‘Ammo-ID’ integration into ballistics workstations and international firearms information-sharing platforms.
• Evaluation of the contribution of additional spectral wavelengths, such as near infra-red, for enhanced computer vision detection and identification.

Expected results

• Library of high-performance EO algorithms and unique ammunition profiles ready for deployment.
• Mobile and desktop software platforms for EO and ammunition detection ready for operational use.
• Assessment reports on algorithm integration and spectral enhancements for computer vision detection.
• Increased engagement and expressed demand from authorities and humanitarian actors.

Expected achievements

• By June 2026, at least 3 authorities have adopted AI and computer vision tools for detection and identification.
• By June 2026, at least 5 humanitarian and human rights actors have integrated AI-CV detection into their operations.
• At least 100 explosive ordnance detection algorithms are produced and fully tested by 2026.
• By 2026, around 1,000 unique ammunition profiles are generated.
• At least 1 mobile EO detection application and 1 ammunition software platform are ready for deployment.
• At least 2 assessment reports on algorithm integration and spectral wavelength applications are delivered.

*Additional countries to be added based on demand