UC2: Obstacle Recogition

Autonomous Underwater Vehicles (AUVs) need reliable and fast obstacle recognition to be able to fly safely though coral reefs and other complex submarine environments. Acoustic sensors are able to look far, but with low resolution. Visual sensors provide high resolution, but cannot look very far.

The idea for the UH-INPA use case in this project stems from the current limitations in surveying coral reefs using an AUV. Coral reefs are delicate structures, that take thousands of years to build. Therefore, the risk of an AUV hitting a coral and destroying it cannot be taken.     

So far, this limitation prohibited autonomous coral reef surveys close to the benthos. The Red Sea, as well as other coral reefs in the world, has a prosperous mesophotic reef environment (depths 30m-120m) that has not been fully explored yet due to limits in human diving to these depths. Nevertheless, this is a very important environment as it is hypothesized that the mesophotic reefs provide a refuge to corals suffering from alleviated ocean temperatures in the shallow depths. Upon the completion of this project UH and INPA will raise funding for extensive surveys of the mesophotic reefs using the new technology.

Current 3D photographic surveys conducted by UH with human divers are
already available online under, with models that have tens of thousands of views. In addition, the coral mapping project has already attracted attention from the largest Israeli online news site and we expect that photographic mappings of the unexplored mesophotic reefs will gain even more attraction and contribute to public education and awareness.

These data will be augmented with the training data generated in WP4 and during the field validations.

Haifa Sea Trials

Photos and Videos: University of Haifa, Matan Yuval