Cognitive robotic system for underwater monitoring and sampling (SENTINEL) (39438)

Project Details


SENTINEL will significantly advance robotics at DTU and will contribute to research, education and innovation in several sectors from computer vision, ocean science and autonomous systems. In the project we will integrate advanced sensors, such as spectral fluorometers and underwater laser cameras, to state-of-the-art underwater vehicles developing machine-learning algorithms for automatic identification of marine features and marine organisms.

- To develop and test autonomous robotic system for water quality analyses and infrastructure inspection in marine environments. The robots must be able to perform automatic identification of targets and selective sampling in the ocean. 
- To implement seafloor and water column navigation, mapping and sensing capabilities of the marine robots with accuracy allowing to use them in confined environments. Robots must be able to operate in coastal zones and harbors.
- To develop automatic classification scheme for real-time characterization of marine features and marine organisms. The system (software and hardware) will be initially customized for operation in Danish waters.
- To contribute to educational frameworks established in REMORA and at DTU Aqua where the students have access to the experimental robotic facilities and use them for real application in marine environment.
- To explore the potential of the system for commercial and industrial applications through knowledge transfer activities with relevant end-users. 

National Institute of Aquatic Resources, DTU Aqua (coordinator)
DTU Electrical Engineering
Maersk Drilling, Denmark
Danish Hydraulic Institute, Denmark
Danish Pelagic Producer Organization, Denmark

The project is funded by Orients Fond.

Research area: Observation Technology
Effective start/end date01/05/201730/04/2019


Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.