Evangelos Boukas

Evangelos Boukas is an Associate Professor in the Department of Electrical and Photonics Engineering at the Technical University of Denmark (DTU), Faculty of Engineering. His research focuses on autonomous robotic systems for marine vessel inspection, confined space navigation, and man overboard detection using UAVs and deep learning techniques.

His core research interests include Deep Learning, Unmanned Aerial Vehicles, Computer Vision, and Robotics, applied to solve challenges in marine robotics and autonomous inspection. He develops uncertainty-aware navigation systems, probabilistic segmentation methods, and reinforcement learning frameworks for real-world deployment in complex environments.

Recent publications demonstrate a strong trend toward GPU-accelerated deep learning for confined space inspection, with emphasis on uncertainty estimation, domain adaptation, and real-time processing. Key application areas include marine vessel classification, ballast tank inspection, and search-and-rescue operations, often leveraging multimodal sensor fusion.

Dr. Boukas actively supervises five PhD students across major research projects:

• Probabilistic Deep Learning for Autonomous Aerial Inspection (2025-2028)
• Aerial Robots Design for Confined Spaces Inspection (2024-2027)
• Active Semantic Segmentation for Confined Spaces using Aerial Robots (2024-2027)
• Autonomous Aerial Robotics for Man Overboard Incidents (2023-2026)
• Visual and Multimodal Perception for Construction Robotics (2023-2026)

He leads the "Perception and Cognition for Autonomous Systems" research group at DTU, which develops advanced perception algorithms for robots operating in challenging environments like ship ballast tanks and marine vessels, contributing to UN Sustainable Development Goals for industry innovation and marine conservation.