Underwater autonomous robots operating 2,000 meters deep will compete against each other and the nature of the ocean in the first round of the Shell Ocean Discovery XPRIZE in late fall 2017.
Virginia DEEP-X, a team of students and faculty led by Professor Dan Stilwell, and 21 other semifinalists will square off in the $7 million global competition. The team is extending its prior accomplishments in subsea robot design, subsea navigation, and collaborative autonomy to explore the vast, unplumbed depths of the ocean.
"The deep ocean is poorly known," said Stillwell. "We have much better maps of the Moon, Mars, and even Venus than our own oceans."
While the technology to conduct high-fidelity surveys already exists, it is slow and extraordinarily expensive, explained Stilwell. XPRIZE seeks technology that is orders of magnitude faster and less expensive than the current state of the art.
DEEP-X is answering the XPRIZE challenge by developing a coordinated team of small, low-cost autonomous underwater vehicles (AUVs).
In the first of two rounds, teams test their entries at a depth of 2,000 meters, mapping at least 20 percent of the 500 km2 competition area at five meters resolution. They will be also be identifying and imaging at least five archeological, biological or geological features at any depth—all within 16 hours.
Round two, the final deep-sea test, will take place at 4,000 meters below the surface, and teams will be required to map at least 250 km2 of the sea-floor.
The DEEP-X team, which is a collaborative project between Virginia Tech and Old Dominion University, has integrated acoustic communication systems into the AUVs so that they can communicate underwater using sound. Communication is necessary for collaboration. The same communication technology also aids navigation, which is the process by which an AUV determines its location. Every time an AUV hears a data packet from another AUV, it is able to infer the range to the other AUV, which is helpful for estimating the location of the AUV.
"One challenge for communication is that the density of seawater varies within the ocean due to the variation of temperature and salinity, among other characteristics, and that changes in density can dramatically affect how sound travels through the ocean. A layer of dense, cold water, for example, might effectively block acoustic communication," said ECE Assistant Professor Ryan Williams, who is participating in the effort. "This makes communication with the surface—and therefore pinpointing robot location—a tricky task."
Autonomous coordination is paramount for a task this ambitious, said Williams. "XPRIZE's ocean mapping effort is arguably harder than putting a robot on the moon because of the parameters they're asking for."
But it's about learning how to think big, said Williams—or, in DEEP-X's case, think deep.