Research water tank

In this 10,000-gallon tank, researchers test robots built to work underwater. Beyond shorting out components, water makes a robot’s other tasks harder too: sensing the environment in murky, dark water where color wavelengths shift; communicating through the water with other robots or a remote operator; and moving against strong currents and waves.
At Michigan, research labs develop computer vision and machine learning algorithms that turn raw sensor data such as sonar into usable information. This enables tasks like autonomous mapping and shipwreck detection. The tank is outfitted with motion-capture cameras that track each robot’s exact position, giving researchers a ground-truth reference for measuring how well those algorithms actually work.
Autonomous systems aren't limited to land or air. Marine robots open up possibilities to explore, transport across and monitor the over 70% of Earth's surface that is water. Water, however, is a very tough environment for robots for several reasons. Water makes it tough to send signals or data. Water distorts the images and video used for vision based sensing. Water is classically wet, making electronics susceptible to damage, and saltwater presents even more challenges in terms of corrosion and durability.
These are all problems that researchers use this water tank to help solve and get robots out on the open water. Listen to a shipwreck expedition led by the Field Robotics Group's Professor Katie Skinner in Alpena, Michigan, at the Thunder Bay National Marine Sanctuary.
So we just deploy the IVER for its first mission of the day. So we deploy it off to the side of the boat. We watch it go to its first waypoint on the surface, and it will dive down for the first leg. The program's lawnmower pattern, which means it will, do eight different legs in kind of a lawnmower fashion for this survey. What we're trying to look at is we know where the shipwreck is. This is a well known wreck. But we're looking for clues that would teach underwater vehicles like ours to be able to find wrecks more efficiently. So we're looking for patterns in the bottom, debris that may have fallen off the ship during the time it was sinking.
Related
- YouTube: A 10,000 gallon robotics research water tank
- YouTube: Artificial Intelligence Trained to Find Shipwrecks