DARPA Chooses Teams For 2007 Autonomous Robot Competition

Oct. 18, 2006
Robots will be on the prowl in an undisclosed location in the United States next November as the Defense Advanced Research Projects Agency presents its Urban Challenge competition. So far, DARPA has chosen 11 teams to participate.

Robots will be on the prowl in an undisclosed location in the United States next November as the Defense Advanced Research Projects Agency presents its Urban Challenge competition. DARPA has chosen 89 teams to participate, drawing talent from top military contractors, universities, electronics companies, research institutions, and even high schools from all over the world.

These teams will build fully autonomous ground vehicles that will have to navigate a mock city environment, complete with traffic and obstacles while obeying all traffic laws and completing simulated military supply missions. The top three teams that complete the 60-mile course in less than six hours will receive trophies.

"The depth and the quality of the field of competitors is a testimony to how far the technology has advanced since the first Grand Challenge in 2004," said Norman Whitaker, DARPA’s Urban Challenge program manager. Much more than a trophy is at stake, though. Military planners expect to make a third of their ground forces autonomous, and they need innovative designs to do it.

DARPA established two tracks for competitors. Teams on Track A will receive up to $1 million in technology development funds. They will have to achieve key technical milestones to receive those funds, though. Also, the government will obtain limited license rights to technologies developed with the funding. Teams on Track B will compete next to teams on Track A, but they won’t receive any funding.

Track A’s 11 teams include industry and academia alike. Autonomous Solutions, the Golem Group, Honeywell Aerospace Advanced Technology, Oshkosh Truck Corp., and Raytheon all will be there. So will teams from the California Institute of Technology, Carnegie Mellon University, Cornell University, the Massachusetts Institute of Technology, Stanford University, and Virginia Tech.

"This is excellent news. It means they have faith in our ability to meet the challenge," said Chris Urmson, Carnegie Mellon Tartan Racing Team technology leader. With General Motors as a premier sponsor, the team is preparing a pair of Chevy Tahoes equipped with computers, sensors, and other gear for autonomous driving.

The teams from Track B hail from around the globe. Princeton University and Palos Verdes High School in California will be represented, as well as Team Helios from Greenfield, Wis., and Sting Racing of Atlanta. Then there's CarOLO, Team AnnieWay, and Team Berlin, all of Germany, Grand Challenge NomadZ of Auckland, New Zealand, and Spring Light of Shanghai, China.

And travel will be the least of the challenges these teams will face. They will have to organize and orchestrate computing elements for mission-level deliberations, mode switching, contingency planning, and navigation-level behaviors. These vehicles also will have to sense, differentiate, and localize objects like other cars and features like lane markings in the environment.

"If you look at the literature on robotic vehicles, it mostly stops at static obstacle avoidance," said Ephrahim Garcia, Cornell team advisor and associate professor of mechanical and aerospace engineering (M&AE). "The specification for this is the California driving manual. Think of every detail you do when you drive, and imagine how you would program that."

Cornell also will be using a Chevy Tahoe in the competition. Cornell team advisor and M&AE associate professor Mark Campbell says the big challenges will be collecting a massive amount of information from video cameras, radar, lidar, and GPS signals and processing that information quickly into a perceptive model of the environment around the vehicle. Probabilistic models, he says, will show what’s happening and predict what will happen next.

Oshkosh will field one of the few teams that will use a modified U.S. military vehicle in the competition. The company will focus its efforts on enhancing the artificial intelligence and sensing systems of its TerraMax robotic truck to compete in an urban environment. TerraMax was one of the five robots that completed the 2005 Grand Challenge, another DARPA competition that had autonomous vehicles attempt a 132-mile course through the deserts of the southwest.

Meanwhile, Raytheon’s team will use a rock-crawling Scorpion vehicle from Preferred Chassis Fabrication that will feature a vehicle management system from Tucson Embedded Systems and navigation and control logic from iRobot. The University of Arizona will provide the on-board modeling and simulation.

The teams will compete in a series of qualifying events, leading up to the finals on November 3, 2007. While they will scramble to complete and test their designs over the next year, they won’t lose sight of the competition’s real goal, which is to save lives.

"Raytheon’s sponsorship of this key technology development is an effort to promote advancement of autonomous vehicles for the battlefield," said Raytheon program manager Russell Mikesell. "Raytheon is committed to developing solutions for minimizing casualties, whether on the battlefield or in complex urban environments."

Sponsored Recommendations

Design AI / ML Applications the Easy Way

March 29, 2024
The AI engineering team provides an overview and project examples of the complete reference solutions based on RA MCUs that are designed for easy integration of AI/ML technology...

Ultra-low Power 48 MHz MCU with Renesas RISC-V CPU Core

March 29, 2024
The industrys first general purpose 32-bit RISC-V MCUs are built with an internally developed CPU core and let embedded system designers develop a wide range of power-conscious...

Asset Management Recognition Demo AI / ML Kit

March 29, 2024
See how to use the scalable Renesas AI Kits to evaluate and test the application examples and develop your own solutions using Reality AI Tools or other available ecosystem and...

RISC-V Unleashes Your Imagination

March 29, 2024
Learn how the R9A02G021 general-purpose MCU with a RISC-V CPU core is designed to address a broad spectrum of energy-efficient, mixed-signal applications.

Comments

To join the conversation, and become an exclusive member of Electronic Design, create an account today!