Using Robots and Artificial Intelligence to Understand the Deep-Sea

Researchers are finding that a combination of AUVs and AI can spot and identify deep-sea species.
Loukia Papadopoulos

In order to best conserve and manage marine biodiversity, scientists need accurate information on what inhabits the seabed. One way to collect such data is with autonomous underwater vehicles (AUV) mounted with cameras.


Processing the data

However, the problems lie with processing the data collected. Now, new research led by the University of Plymouth finds artificial intelligence (AI) could help with the task.

Marine scientists and robotics experts tested the effectiveness of a computer vision (CV) system in identifying sea creatures and found it be around 80% accurate. The system could even be 93% accurate if enough data is used to train the algorithm.

"Autonomous vehicles are a vital tool for surveying large areas of the seabed deeper than 60m (the depth most divers can reach). But we are currently not able to manually analyse more than a fraction of that data. This research shows AI is a promising tool but our AI classifier would still be wrong one out of five times, if it was used to identify animals in our images," said Ph.D. student Nils Piechaud, lead author on the study.

"This makes it an important step forward in dealing with the huge amounts of data being generated from the ocean floor, and shows it can help speed up analysis when used for detecting some species. But we are not at the point of considering it a suitable complete replacement for humans at this stage."


The study saw one of the UK's national AUVs, called the Autosub6000, collect more than 150,000 images in a single dive from around 1200m beneath the ocean surface on the north-east side of Rockall Bank, in the North East Atlantic. Researchers then analyzed 1,200 of these images manually containing 40,000 individuals of 110 different kinds of animals.

Most Popular

They then used Google's Tensorflow, an open access library, to teach a pre-trained Convolutional Neural Network (CNN) to identify the species found in the AUV images. They found the method had an 80% rate of accuracy whereas humans perform within a range of 50 to 95%.

"Most of our planet is deep sea, a vast area in which we have equally large knowledge gaps. With increasing pressures on the marine environment including climate change, it is imperative that we understand our oceans and the habitats and species found within them. In the age of robotic and autonomous vehicles, big data, and global open research, the development of AI tools with the potential to help speed up our acquisition of knowledge is an exciting and much-needed advance," said Dr. Kerry Howell, Associate Professor in Marine Ecology and Principal Investigator for the Deep Links project.

The new study is published in Marine Ecology Progress Series