Atlas Robots Released To DARPA Challenge Teams

Friday, August 23, 2013

Boston Dynamics Atlas Robot

DARPA Robotics Challenge
The qualifying Track B teams from the DARPA Robotics Challenge have recently received their ATLAS humanoid robots.  The teams, who got to this stage of the competition by proving their strategies on simulated robots, now get to work on the real thing.

The 330 pound Boston Dynamics-made Atlas robots have been delivered to the DARPA Robotics Challenge Track B teams.

Atlas is intended to be a first-responder or rescue robot, working in environments too dangerous for humans.

While Atlas may have the robust body for the task the DARPA challenge participants are working hard to deliver it's brains.

The agency is currently sponsoring a $2 million competition, with seven institutions vying to program Atlas for a real-world test responding to a disaster, scheduled for later this year.

“We’re in the playoffs,” said Michael A. Gennert, director of the robotics engineering program at Worcester Polytechnic Institute (WPI), one of the teams in the competition.

Also running a team is the Massachusetts Institute of Technology in Cambridge. “We’re in this to win it,” said Seth Teller, who leads an Atlas programming team at MIT’s Computer Science and Artificial Intelligence Lab (CSAIL).

The intra-state competitors underscore how Massachusetts has risen to prominence as a center for robotics research. In addition to Boston Dynamics, other firms from the state include Bedford-based iRobot Corp., maker of the popular Roomba robotic vacuum cleaner. IRobot also designed a mechanical hand for Atlas.

DRC Atlas Robot

DARPA and Boston Dynamics launched Atlas following the 2011 disaster at the Fukushima nuclear power station in Japan. Plant managers used tracked robots from iRobot to remotely inspect areas too radioactive for humans. But these machines could not reach parts of the facility that were accessible on foot.

“Those robots couldn’t really do much more than look around,” said Teller.

Boston Dynamics cofounder Marc Raibert believes that for moving through rocky or rubble-strewn terrain, robots with legs are the best solution.

Related articles
“They can go places where cars and wheeled and tracked vehicles can’t go,” Raibert said.

“Many of the places where disasters might occur are places that are designed for people,” Raibert told the Boston Globe. “People can fit in there and maneuver through them.”

So too are the intentions with Atlas, using the same tools as human first responders. A humanoid robot could climb into a car and drive itself to the disaster scene. Once there, it could open doors, climb ladders, turn valves or throw switches, just like a person.

In an early phase of the competition, 26 teams from eight countries wrote software to control a computer simulation of Atlas.

A consortium of Florida universities took first place, but WPI was runner-up and MIT was third. The two Massachusetts schools and five other teams will each get a real Atlas robot and financial support from DARPA.

In December, they’ll go to Homestead, Florids, to compete against each other, and against at least six Track A teams building their own robots from scratch.

“We’re moving from a simulated world where everything is neat and clean to a real physical world, which is apt to be messier,” said Teller.

The top-scoring teams will qualify for the finals, to be held in December of 2014. By then, Boston Dynamics expects to be able to provide a generator module for each Atlas, which will eliminate the currently used tether and allow the robot to walk freely. And again, the machines, controlled only by radio commands and built-in software, will be given the same eight tests, with $2 million for the top performing team.

Despite the high stakes, the various schools have worked together, sharing ideas on solving tough problems. And the WPI and MIT teams seem determined to keep it classy.

“Those guys are great,” R.J. Linton, project director for WPI’s Atlas team, said of his Cambridge rivals.

MIT professor Russ Tedrake praised the Worcester team for beating his squad in the preliminary round. “I think they did an amazing job.”

For Linton, it’s not a matter of money or bragging rights.

“What this is about is pushing technology and advancing the state of the art so we can save people’s lives,” he said.

SOURCE  Boston Globe, IEEE Spectrum, Top Image - Essdras M Suarez/Boston Globe

By 33rd SquareSubscribe to 33rd Square