'Team Case' collaborates with ENSCO, Inc. to engineer autonomous, biologically-inspired robotic vehicle—"DEXTER"—in pursuit of $2 million first prize.
Everyone knows that teaching teenagers how to drive is a difficult task. Teaching a car how to drive—by itself—is quite another.
And teaching the car to perform all kinds of driving behaviors independently that most human beings know instinctively might be the equivalent of teaching someone with two left feet how to dance.
But a team of about 50 undergraduate and graduate engineering students and faculty from Case Western Reserve University is rising to that challenge—all in pursuit of a $2 million first prize from the United States Department of Defense. "Team Case" is one of 85 national and international teams that will enter the 2007 Defense Advanced Research Projects Agency (DARPA) Urban Challenge for autonomous robotic ground vehicles. The Urban Challenge, scheduled for November 3, 2007, is a successor to the DARPA Desert Challenges of 2004 and 2005. The top three finishers receive $2 million, $1 million and $500,000 awards and gain the satisfaction and prestige of knowing that they helped develop technology that may someday protect the lives of American men and women on the battlefield.
The Urban Challenge is designed to develop autonomous driving skills for city streets, eventually leading to cars that drive themselves and meeting a Congressional mandate by 2015 that one-third of operational ground combat vehicles are unmanned. The contest will feature the vehicles conducting simulated military supply missions safely and effectively in a mock urban area. At an as yet undisclosed location in the southwestern U.S., robotic vehicles will attempt to complete a 60-mile course through traffic within six hours, operating under their own computer-based control. To succeed, vehicles must obey traffic laws while merging into moving traffic, navigating traffic circles, negotiating busy intersections and avoiding obstacles.
Team Case, however, a first-time entrant, still must qualify for the race.
"We still have to qualify, and we're confident we will," said Andy Allen, 24, an engineering doctoral student who is the project's technical lead for the car's situational awareness, making him responsible for collecting information from processed sensor data and making strategic decisions from that data. "Part of our advantage is that we already have a car built and we don't have to start from the ground up. But it still has to be modified to intelligently navigate in an urban environment."
To qualify, the team will have to go through a few hoops, Allen says. First, in order to get to the National Qualifying Event (NQE) in October, the team must submit to DARPA by April a video of the car performing basic navigation and basic interaction with other vehicles. DARPA will then review all the videos and grant site visits to a handful of teams. While on site visits, representatives of DARPA will watch vehicle interaction on a test track.
"We will be allowed to compete in the NQE if DARPA likes how our robot performs," Allen said. "I'm confident that we're going to do it."
Through a partnership with ENSCO, Inc., a well-established engineering consulting and contracting firm based in Falls Church, Va., Team Case is modifying DEXTER, the autonomous robot designed originally by ENSCO that placed 6th in the Desert Challenge in 2005 and has been lent to Case. The car will be a novel combination of a road-tested vehicle and sensors, paired with a biologically-inspired approach to environmental perception, humanistic driving instincts and vehicle control.
DEXTER's desert performance and perfect run at the 2005 National Qualifying Event proves that it can not only handle the abuse of severe terrain but also navigate with the precision required in a city, according to Wyatt Newman, faculty leader of Team Case and professor of electrical engineering and computer science at Case.
Since Case is among the leaders in the study and application of biologically-inspired robotics, it will bring a behavioral programming approach to planning and vehicle control. The DARPA Urban Challenge is far more complex than the Desert Challenge, mimicking more intelligent human driving behaviors, including situational awareness, to improve responsiveness and adaptability within a fluid and dynamic environment.
"Using special software, DEXTER is being taught to utilize human-inspired thinking to choose behaviors based on safe driving ability and legality, important pillars in DEXTER's operation," said Bradley Hughes, a 19-year-old freshman engineering major from Pittsburgh, Pa., who is a member of the project's vehicle control group and is working on increasing the accuracy of the car's pose awareness from GPS and inertial sensors.
In keeping with the university's emphasis on experiential learning, Newman says that the experience gained by Case students in such a labor-intensive project will carry them into their professional lives.
"The Urban Challenge constitutes a level of complexity and excitement that provides students and faculty with practical goals and tests of their skills in a demanding environment," said Newman. "The opportunity for our students to work on practical problems using leading-edge technology is a learning experience they'll be able to take with them and apply to their professional lives after they graduate."
Case Western Reserve University is committed to the free exchange of ideas, reasoned debate and intellectual dialogue. Speakers and scholars with a diversity of opinions and perspectives are invited to the campus to provide the community with important points of view, some of which may be deemed controversial. The views and opinions of those invited to speak on the campus do not necessarily reflect the views of the university administration or any other segment of the university community.