Skip to content


Nataliya Strokina
Local flow sensing for underwater robots
On 2015-04-02 11:00 at G205
Due to emerging applications of autonomous submersible systems, for example, in
environmental monitoring and sea border control, the underwater navigation has
become an important problem.

Successful approaches in autonomous vehicles, e.g., using the global positioning
system (GPS) and egocentric sensor measurements (vision and sonar), can be
adopted, but conditions such as the lack of map points in open water, diving in
underwater caves, and limited visibility justify investigating other available
sensor modalities. 

In this project we investigate signals produced by an array of pressure sensors
mimicking the biological lateral line of fishes. We introduce a fish-shaped
artificial lateral line sensor, calibration of the sensor array signal, and
propose signal processing methods for the important low-level tasks required for
flow-based underwater navigation: bulk flow angle and flow velocity estimation.
In our experiments we demonstrate the feasibility of an artificial lateral line
sensor to provide accurate and robust ``local flow feature''
Back to the list