Sonar system helps underwater robots navigate cloudy seafloors

Remotely operated underwater vehicles can lose visibility when they land on the seafloor or dig into sand, stirring up sediment that onboard cameras cannot see through. A system developed by Amy Phung, SM ’23, PhD ’26, and Richard Camilli, SM ’00, PhD ’03, of the Woods Hole Oceanographic Institution gives these vehicles a way to continue operating instead of waiting for the water to clear.

The method uses sonar to rapidly map nearby surroundings, providing enough spatial awareness for a vehicle to move safely toward objects even in cloudy water. Cameras can then inspect targets at close range once the vehicle is properly positioned. To enable real-time mapping, the researchers paired sonar with an image-matching algorithm developed by researchers in France that estimates the relative depth of each pixel in a 2D scene.

Camilli compared the approach to navigating a dark china shop to find a specific coffee mug without knocking anything over. Potential uses include scientific exploration, underwater construction and maintenance, and handling unexploded undersea mines.