MIT model explains how the brain can learn novel tasks while still remembering what it has already learned.
In a step forward for research in the seas, MIT scientists have successfully launched an autonomous underwater vehicle (AUV) that is cheap enough, adaptable enough and small enough to run oceanographic missions on real-life, non-military research budgets.
The launch is important because it was from an oceanographic vessel and in Antarctica, where conditions are demanding and logistics difficult. (Until now the AUV, named Odyssey, had only been deployed in the Charles River and off Nahant.)
The launch late last month was conducted by James G. Bellingham, manager of the MIT Sea Grant Underwater Vehicles Laboratory where Odyssey was designed and built, and postdoctoral fellow Donald Atwood.
Odyssey was lowered by winch into the water from the RV Nathaniel B. Palmer, an oceanographic research vessel belonging to the National Science Foundation. Drs. Bellingham and Atwood, waiting alongside the Nathaniel B. Palmer in an inflatable boat, connected a computer to the AUV, fed it its computer instructions for the day, and sent it on its missions.
Odyssey made multiple runs on three different days in both good weather (for Antarctica) and in seas building to 20-foot swells. According to Dr. Bellingham the launchings provided valuable lessons for future development-among them, how to spot the small vehicle once it resurfaces.
"People on the ship [the Palmer] could easily see the vehicle, while we in the Zodiac [the inflatable], being much closer, couldn't," Dr. Bellingham said. "With what we know now, we can make the vehicle easier to relocate by at least a factor of five .
"We also learned very practical lessons regarding using the full potential of the oceanographic vessel and have some ideas for efficient strategies." In addition, Dr. Bellingham said, "We also found that Odyssey attracts birds, providing yet another research tool.
Odyssey is a powerful oceanographic research tool. It is capable of diving to about 19,000 feet, deep enough to cover about 97 percent of the ocean floor. On this mission the vehicle employed sonar and a video camera; the laboratory plans to add additional capabilities, including temperature and chemical sensors.
Further, at about 360 pounds, a range of just under 170 miles, and a component cost of under $50,000, Odyssey is one sixth the weight, has twice the range and is a fraction of the cost of any other deep-diving AUV in the world.
AUVs in general have a significant advantage over the better-known remotely operated underwater vehicles (ROVs), such as the Woods Hole Oceanographic Institution's Jason: AUVs have no tether.
All power and all instructions-in the form of computer programs-are stored aboard. This means that AUVs do not have to be continuously tended by a surface ship and they do not run into the often disabling problems caused by miles of cable pushed around by oceanic forces. For example, ROVs are subject to loss when cables break and only special vessels can launch them. (The mission of Dante, a tethered, eight-legged robot designed to crawl into Antarctica's Mt. Erebus volcano, was ruined by cable problems this January.)
The introduction of small AUVs is predicted to vastly extend the ability of oceanographers to study the underwater world. AUVs can go where humans can't, or can't afford to-from under ice in the Arctic to over hydrothermal vents in the deep sea. Until now, visits to such locales had been achieved only by large, expensive vehicles-too large, and too expensive, for many researchers.
But now Odyssey has proven that it can go out in Antarctic seas, do a computer-programmed mission, and come back to its home ship safe and sound.
A version of this article appeared in the February 10, 1993 issue of MIT Tech Talk (Volume 37, Number 22).