Team Michigan Selected as Finalist in MAGIC 2010 Robot Challenge   Bookmark and Share




Team Michigan, a research team led by Assistant Professor Edwin Olson and his students, has been selected as one of six finalist teams in the Multi Autonomous Ground-Robotic International Challenge (MAGIC) 2010 competition. The team, which previously qualified for funding in November 2009 as one of 12 semi-finalist teams, will now receive additional funding and advance to the MAGIC Grand Challenge Event this November 8–13 in South Australia.

The MAGIC Challenge is jointly sponsored by the Australian and US Departments of Defense. The competition challenges robotics development teams from around the world to develop next-generation fully autonomous ground vehicle systems that can be deployed effectively in military operations and civilian emergency situations. During the competition, teams of robots are expected to explore indoor and outdoor environments, identify and track humans, and detect objects of interest. The team that places first in the November Grand Challenge will receive a US$750,000 prize in recognition of their significant accomplishment and to fund additional research in this area.

Team Michigan was selected as a finalist in the MAGIC Challenge after competition organizers conducted site visits of the 12 semi-finalist teams over the course of the summer. In announcing the finalists, Dr. Grace Bochenek, U.S. Army Tank Automotive Research, Development and Engineering Center director said, "All the judges were impressed with the magnitude of submissions from the teams. These finalists have survived a rigorous assessment and elimination process against six other semi-finalist teams."

Team Michigan: Robots and More Robots... with All the Right Stuff

Team Michigan's approach to the MAGIC Challenge has been to develop and deploy a large field of low-cost robots from mostly commodity components, and to leverage cognitive science research at Michigan to imbue this robotic team with as much autonomy and collaborative intelligence as possible. The result is a team of reliable robots that can work together seamlessly to execute missions and which can reconstitute to complete missions as some robots are "taken out" by judges mid-stream in competition.

This remarkable robotic intelligence is based on research into machine perception, autonomy, and learning that takes place in Prof. Olson's APRIL Laboratory. It also draws on decades of research led by Prof. John Laird into the creation of an overarching artificial intelligence architecture with cognitive decision making and reaction capabilities called Soar. Soar Technology, a cognitive science company founded by Prof. Laird, is also a member of Team Michigan.

Read More:

MAGIC Finalists Announcement
U-M News Service Press Release
MAGIC Challenge Website
MAGIC Challenge Facebook page
APRIL Robotics Laboratory Website

 
Graduate student Ryan Morton speaks about the goals of the site visit


Video from the recent U-M site visit




Below are photos from the recent U-M site visit. To view larger photos, just click on the thumbnail image of interest.


Posted: July 27, 2010