Classification of sonar data with deep learning

The aim of this project is to develop and test algorithms for classification of sonar data. The project will use Deep Learning as a tool to make autonomous underwater system more robust when unknown objects are detected.
The first part of this sub-project addresses the problem of Deep Learning of sonar data, and the second part aims to use the algorithms from the first part to enhance the performance of autonomous vehicles performing different tasks.

Per Johansson, Project manager Saab Underwatersystems  – Project leader

Louise Fuchs – Industrial PhD-student Saab Dynamics