IMI at University of Bath Develops AI to Map Marine Environments

Working with SEA Ltd., IMI has developed an Artificial Intelligence (AI) algorithm which could improve how we monitor marine territories.

The developed AI model can automatically classify underwater environments directly from sonar measurements.

The UK's vast marine territories are monitored by the Royal Navy using high tech sonar equipment. Sonar systems emit and receive acoustic signals which can be analyzed to detect underwater objects and map subsea environments.

The Defense and Security Accelerator ¹ (DASA), a part of the Ministry of Defense (MoD), contracted the IMI and Systems Engineering & Assessment Ltd. (SEA) to develop an AI algorithm capable of automatically classifying underwater environments directly from sonar measurements.

Developing AI algorithms to classify underwater environments

Underwater environments vary hugely in terms of water temperature, salinity and depth as well as seabed slope and composition, all of which affect sonar. IMI and SEA first analyzed the many characteristics of underwater environments and classified them into different types.

They then reviewed various AI techniques to determine the most effective classification approach. The selected method (Probabilistic Generative Modelling) was then adopted as a framework to develop three different AI algorithms for identifying underwater environments.

A Probabilistic Principal Component Analysis (PPCA) model proved to be the basis for the most effective algorithm. They then developed the model further through rigorous experimentation to achieve the highest possible classification accuracy.

Classification accuracy of up to 96%

After developing the AI algorithm, the performance was tested on a wide range of simulated acoustic data representative of a broad spectrum of underwater environments.

The tests demonstrated that the PPCA algorithm can classify underwater environments from simulated sonar measurements with an average accuracy of 93%. Classification accuracy remained high even when we used a short spatial interval of the test data, which is promising for the practical use of the technique.

An alternative Latent Variable Gaussian Process (LVGP) model also showed strong performance and enabled the achievement of an even higher classification accuracy of 96%.

‘This project exceeded all our expectations for AI algorithms applied to the complexity of sonar in the underwater environment. We look forward to continuing our collaboration with the IMI following positive feedback from the MoD.’ said Marcus DonnellyTechnical Lead, Environmental Data Science, SEA Ltd.

DASA finds and funds exploitable innovation to support UK defense and security quickly and effectively, and support UK prosperity. DASA’s vision is for the UK to maintain its strategic advantage over its adversaries through the most innovative defense and security capabilities in the world. DASA is a cross-Government organization, launched in December 2016 by the Secretary of State for Defense.