AI & Robotics
National
AEOLUS
AI-driven Efficient Object Detection for Low-power Unmanned Surveillance.
AEOLUS develops energy-efficient, AI-based object-detection methods that bring real-time computer vision to low-power unmanned platforms — extending surveillance capability where size, weight and power budgets are tightly constrained.

overview
Seeing more with less power.
Modern surveillance increasingly relies on small unmanned platforms — but advanced object detection is computationally expensive and quickly drains constrained power budgets. AEOLUS tackles this gap directly, designing AI models and pipelines that deliver accurate, real-time detection at a fraction of the energy cost.
Through model optimisation, efficient inference and hardware-aware design, the project extends the operational envelope of low-power unmanned systems — enabling longer missions, lighter payloads and dependable autonomy in the field.
Objectives & outcomes
- Develop energy-efficient AI object-detection models for embedded platforms.
- Achieve real-time inference within strict size, weight and power limits.
- Validate detection accuracy and robustness in realistic field conditions.
- Deliver a deployable prototype pipeline for low-power unmanned systems.
Project facts
Programme
RESTART — Research & Innovation Foundation (RIF)
CYRIC role
R&D Partner
Duration
2024 – 2026
Total budget
€0.8M (placeholder)
Partners
CYRIC · University of Cyprus · industry partner
Programme
ACtive
Dissemination
Resources & downloads
Posters
Download project posters and visual materials.
Publications
Access research papers and publications.
Leaflets
Download informational leaflets and brochures.
Conferences
View conference presentations and materials



