Smart Takes

The robots replacing dolphins and humans: Scouting for mines

The robots replacing dolphins and humans: Scouting for mines

Posting in Design

The U.S. Navy may one day be able to retire their human and animal divers as robotic research steps up.

In order to find and destroy underwater mines attached to ship hulls, military forces have employed human divers to undertake the time-consuming and sometimes dangerous task.

However, the U.S. Navy have also trained dolphins and seal lions to search for these types of bombs -- able to cover large areas in short spaces of time, but costly to raise and unpredictable in their performance.

The Massachusetts Institute of Technology (MIT) reports that the Navy's research department has been engineering robots to take over the roles of minesweeping and other dangerous underwater missions in the last few years. The aim is to develop a completely autonomous machine that can navigate and map murky underwater environments -- without prior knowledge of them -- and detect mines the size of an iPod.

However, this dream has yet to be realized. In pursuit of the ideal, Franz Hover, the Finmeccanica Career Development Associate Professor in the Department of Mechanical Engineering, and graduate student Brendan Englot have designed algorithms that improve robotic navigation and detection capabilities.

Using these new, complex algorithms, a robot is able to swim around a ship's hull and detect complex structures including propellers and shafts. With further refining, the team aim to achieve a resolution fine enough to detect 10-centimeter mines attached to ship hulls.

The research is documented in a paper to appear in the International Journal of Robotics Research.

"A mine this small may not sink the vessel or cause loss of life, but if it bends the shaft, or damages the bearing, you still have a big problem," Hover says. "The ability to ensure that the bottom of the boat doesn't have a mine attached to it is really critical to vessel security today."

The algorithms are being used to program a robot dubbed the Hovering Autonomous Underwater Vehicle (HAUV), which was originally part of MIT's Sea Grant program. However, placing the software from theory to practice has not been without its complications. Hover said:

"It's not enough to just view it from a safe distance. The vehicle has to go in and fly through the propellers and the rudders, trying to sweep everything, usually with short-range sensors that have a limited field of view."

The challenge of developing these sophisticated algorithms was approach in two stages. In the first stage, the team made a robot approach a ship's hull; using a sonar camera to ping back signals into what they describe as a "grainy point cloud".

The point cloud operates as a map for the robot. However, sonar technology would not be able to tell the machine where the ship's structure begins or ends -- so in order to translate the "mist", Hover's team added computer-graphics algorithms to their sonar data, which created a 3D mesh model of the robot's surroundings.

In the second stage, they changed the boundary that the robot was allowed to approach from meters to centimeters. The idea is that a robot should be able to cover every point in the mesh -- and so each point was altered to be ten centimeters apart, which is small enough to detect small mines.

The team has conducted field tests with the algorithms, using underwater models of two vessels; a 183-meter military support ship and an 82-meter cutter.

"The goal is to be competitive with divers in speed and efficiency, covering every square inch of a ship," Englot says. "We think we’re close."

This research is supported by the Office of Naval Research.

For more information, view the video below:

Image credit: MIT/ Franz Hover/ Brendan Englot

Share this

Charlie Osborne

Contributing Editor

Charlie Osborne is a freelance journalist and photographer based in London. In addition to SmartPlanet, she also writes for business technology website ZDNet and consumer technology site CNET. She holds a degree in medical anthropology from the University of Kent. Follow her on Twitter. Disclosure