Neuromorphic Event-Driven Vision for Multiple Defence Applications.
Conventional frame-based cameras struggle with fast targets, poor light and cluttered scenes. Event-driven sensors work differently: each pixel reports only what changes, which cuts latency, power consumption and data volume. NEVMA turns that principle into a neuromorphic vision system for modern defence operations and validates it in prototypes for air, land and maritime use.
The European Commission selected NEVMA for funding under the European Defence Fund (EDF) 2025 call for development actions dedicated to SMEs.
Project details
| Full title | Neuromorphic Event-Driven Vision for Multiple Defence Applications |
| Acronym | NEVMA |
| Grant agreement | 101304923 |
| Programme | European Defence Fund (EDF) 2025 |
| Call | Development actions dedicated to SMEs |
| Topic | Non-thematic development actions by SMEs (EDF-2025-LS-DA-SME-NT) |
| Duration | 36 months |
| Type of activities | Studies, design, system prototyping, testing |
| Estimated total cost | € 5,781,668.83 |
| Maximum EU contribution | € 5,028,348.79 |
| Coordinator | MILTECH HELLAS SA (Greece) |
Where NEVMA innovates
The consortium concentrates its work on three capability gaps:
- High-speed threat detection and classification. Event-driven sensors capture microsecond-scale motion, so the system tracks and classifies fast, small or erratic targets that frame-based cameras blur or miss.
- Autonomous navigation in GNSS-denied environments. Visual odometry on event data keeps platforms oriented where satellite positioning fails, whether jamming, spoofing or terrain causes the loss.
- Low-light and obscured-environment surveillance. The high dynamic range of neuromorphic sensors holds up at night and through smoke, dust, fog and glare.
From the laboratory to near-operational scenarios
NEVMA advances the technology from TRL 4 to TRL 7 over a three-year work plan and delivers integrated prototypes that partners validate in near-operational scenarios:
- UAV interception: detection, tracking and classification of drones at speed and at range.
- GNSS-free navigation: position and attitude estimation for platforms that operate without satellite signals.
- Coastal surveillance: persistent maritime border monitoring under low light and adverse weather.
Each demonstrator combines neuromorphic sensing, embedded processing and AI-based perception in a single prototype chain, so end users assess a complete capability rather than isolated components.
Consortium
| Partner | Country |
|---|---|
| MILTECH HELLAS SA ANTIPROSOPEIES ILEKTRONIKON OPTIKON KAI MICHANOLOGIKON EIDON AE (Coordinator) | Greece |
| BENE CERCETARE SRL | Romania |
| DEFSECINTEL SOLUTIONS OÜ | Estonia |
| INFILI TECHNOLOGIES SOCIETE ANONYME | Greece |
| NEVROMORFIKA EOOD | Bulgaria |
| TELETEL TECHNOLOGIA AERODIASTIMIKIS KAI AMINAS | Greece |
Six SMEs from four Member States form the consortium, which pairs defence system integrators with specialists in neuromorphic hardware, artificial intelligence and secure communications.
The role of BENE Cercetare
BENE Cercetare contributes its experience in sensor integration, edge processing and secure data communication to the NEVMA prototypes. The team also supports test and validation activities and leads dissemination, communication and exploitation, so that project results reach defence stakeholders, the research community and the wider European defence technological and industrial base.
Funding
Funded by the European Union under the European Defence Fund (EDF), grant agreement No 101304923. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or the European Commission. Neither the European Union nor the granting authority can be held responsible for them.