Advertise

Home > Aircraft technology > Robot Research Focuses On Versatility

Robot Research Focuses On Versatility

In 2001, the U.S. Congress handed the Pentagon its marching orders for unmanned technologies: One-third of the ground combat vehicle fleet was to be remotely operated by 2015.

A decade later, it is obvious that barring a major expansion of innovation the goal won’t be met. There are more than 2,000 ground robots in Afghanistan, the majority of which are used by explosive ordnance teams. But as essential as those bots are, there is much more going on in robotics, from lightweight “throwbots” to swarms of robots controlled by a few users and experiments with cellular-based control networks.

Speaking at the Association for Unmanned Vehicle Systems International (AUVSI) symposium in Washington last month, the U.S. Army’s chief scientist, Scott Fish, said that while demonstrations of semi-autonomous unmanned ground vehicles (UGVs) have abounded over the past five years, leaders he has spoken with still don’t know when they can deliver on the promise of greater autonomy. He said the Army needs to do a cost/benefit analysis of how much robots cost, what it would take to train soldiers to use and repair them, and how many the service needs and to what end before it can move forward with plans to integrate UGVs into the force. “We need to go back and look at what we want to do autonomously,” he said, voicing frustration that the work hasn’t been as organized as it could have been. “We’ve got to change what we’re doing—it’s not working.” (For an interview with Fish, see DTI February, p. 52.)

Right now—as James Overholt, chief robotics scientist of the Army’s Tank Automotive Research, Development and Engineering Center (Tardec), said during his talk—the unmanned systems in use are mission enablers, as opposed to independently operating systems that can be considered partners. “We need to go through a social and cultural change” in the military, and in society, Overholt said, “to get to where these systems are accepted as mission enhancers that allow a safe standoff distance for their human operators, but can also perform missions themselves, allowing soldiers to do something else.”

Rob Maline of the Pentagon’s Joint Ground Robotics Exercise made similar points in his address, noting that the military doesn’t “have a Defense Department ground robotics vision” yet, but his shop is working on it and hopes to unveil one this spring.

Currently, the ratio of unmanned systems to human operators is still 1:1, which makes the use of robots in combat a personnel-heavy affair. But “we’re trying to change this paradigm,” says Edwin Olson, a professor of computer science and engineering at the University of Michigan. “We want to have a team of robots that is operated by one or two operators.”

The proof that such technology is not only available but ready for use comes from a military robot competition held last November in Australia. The competition, called the Multi Autonomous Ground-robotic International Challenge, or Magic, was organized jointly by Australia’s Defense Science and Technology Organization and Tardec.

Olson’s University of Michigan-led team won the $750,000 cash prize using 14 robots that mapped out an obstacle course littered with improvised explosive devices.

“The notion there is that one of the more dangerous missions a soldier has to undertake is to go into an unknown building and scout through the building to see if there is anything that’s a danger to the people going in,” says Overholt. “I would like to facilitate that with robots.”

Robot competitions are nothing new, even for the military. The Defense Advanced Research Projects Agency (Darpa), for example, held a series of Grand Challenge competitions to demonstrate the ability of cars to drive autonomously. “The Magic competition was different fundamentally, first of all, because it had multiple robots working together, but also because there was a human in the loop,” says Olson, who was also a student technical lead on MIT’s Urban Challenge team, which placed fourth in the Darpa competition.

Recent Entries

Leave a Reply

Improve the web with Nofollow Reciprocity.