Electronic Design

Robotic Cars Get Street Smart

This November, Mike Montemerlo's Volkswagen Passat wagon will drive a 60-mile trek through an urban landscape located somewhere in the western U.S. But Montemerlo won't be sitting behind the wheel, nor will anyone else. That's because Montemerlo's Passat happens to be a special robot model, custom-developed by the Stanford University Racing Team (Fig. 1).

Along with robotic-vehicle researchers at other institutions worldwide, the Stanford team will compete in the DARPA (Defense Advanced Research Projects Agency) Urban Challenge, an event sponsored by the U.S. Department of Defense. DARPA's mission is to maintain the military's technological edge and to prevent new technology from jeopardizing national security. To meet these goals, the agency sponsors revolutionary, high-payoff research ventures such as the Urban Challenge that can lead to innovative new military capabilities as well as leading-edge consumer or business products and services.

The Urban Challenge's roots lay in a 2001 Congressional mandate that requires one-third of the nation's combat ground vehicles to be unmanned by 2015. To help reach this goal, Congress authorized the Department of Defense and DARPA to award funding and cash prizes to Urban Challenge participants. Eleven teams received seed money of up to $1 million for development while another 78 are funded either by themselves or by corporate sponsors. The competition's top three finishers can look forward to receiving cash prizes of $2 million, $1 million, and $500,000, respectively.

Self Control
The Urban Challenge is designed to test the safety of autonomous vehicles as they interact with other vehicles and the local environment. Come this November, dozens of vehicles will be placed onto a course that simulates real-world city streets.

"All of the participants' vehicles will be on the course at the same time, with some additional traffic vehicles that we provide," says Norman Whitaker, the DARPA's Urban Challenge program manager. "It really tests lots of different dimensions of the technology... both their ability to sense and react correctly in traffic as well as their ability to be able to adapt to changing \[road\] conditions."

To succeed in the Urban Challenge, competing teams' entries must perform like vehicles controlled by human drivers and safely conduct simulated battlefield supply missions on the 60-mile course. The vehicles must obey strict rules while merging into traffic, navigating traffic circles, negotiating busy intersections, and avoiding obstacles before finishing under six hours. The competition's location will be announced in August.

Building A Robot Car
The strongest Urban Challenge competitors are teamed with major automakers and other sponsors. Volkswagen, in Stanford's case, provides research expertise and funding. "Volkswagen does all the retrofitting of the vehicle and all the vehicle electronics," says Montemerlo, the team's software lead and a senior research engineer in the Stanford Artificial Intelligence Lab. "At Stanford, we concentrate on what we do best, which is the software and sensing."

Stanford's robo-Passat, nicknamed "Junior," is like a NASCAR racer in that the vehicle is so highly modified it bears little more than a passing exterior resemblance to its street counterpart. Junior's steering, throttle, and brakes were all modified by engineers at the Volkswagen of America Electronics Research Laboratory in Palo Alto, Calif., to be completely computer-controllable (Fig. 2). The engineers also created custom mountings for a bevy of sophisticated sensors.

Junior's standard equipment includes a range-finding laser array that spins to provide a 360° 3D view of the surrounding environment, nearly in real-time. Six video cameras that accompany the array "see" all around the car. Junior also uses bumper-mounted lasers, radar, GPS receivers, and inertial navigation hardware to collect data about where it is and what's nearby.

According to Montemerlo, developing and installing the control and monitoring hardware was the easy part. Teaching Junior how to safely drive itself is proving to be far more tricky, requiring programmers not only to instruct the robot on how to control itself, but also how to anticipate changes in traffic conditions. "You have to make models of other vehicles and predict how you think they're going to behave," Montemerlo says.

Junior's custom-coded software modules include a planner (for making decisions and choosing routes), a mapper (which transforms sensor readings into environment understanding), a localizer (which refines the GPS position by visual observations), and a controller (which actuates planner decisions). The entire system runs on rack-mounted servers equipped with Intel Core 2 Duo processors. Data is processed from the vehicle's instruments as frequently as 200 times per second, Montemerlo notes.

But even with all of its cutting-edge hardware and software support, Junior remains at a cognitive disadvantage compared to even a novice human driver. "The robot actually has a lot less information about what other vehicles might be doing," Montemerlo says. "A robot certainly won't be able to interpret other cars' turn signals, for instance." Nonetheless, he's hopeful that Junior will perform well this November. "We do our best to train for the worst possibility and hope that it works out," he says.

Victor Tango
Also competing for Urban Challenge gold is Virginia Tech and its Victor Tango racing team. The school's robotic vehicle, named Odin, is based on a hybrid Ford Escape (one of two donated by the automaker). The rolling robot incorporates several urban-oriented autonomous vehicle technologies, including traffic behavior modeling, route planning, autonomous parking, road detection, and vehicle passing. Like Stanford's Junior, Odin uses a variety of sensor technologies, including computer vision, laser range-finders, differential GPS, and inertial measurement navigation.

Alfred Wicks, an associate professor of mechanical engineering at the school, believes that a combination of computer vision and laser range-finding is crucial for spotting and avoiding obstacles. "The vision system works like peoples' eyes. It collects data and it's processed," he says. "To supplement that vision system, the laser range-finders give range and angles." The overall system helps Odin's computer distinguish between moving cars and static objects.

Wicks believes that events like the Urban Challenge give engineering students a taste of the excitement and recognition usually reserved for athletes. "Students get to go out and compete, not chasing a little piece of pigskin around the football field, but competing in technology," he says. DARPA has given Victor Tango the number 32 to honor the people who lost their lives in last spring's campus shooting tragedy.

Road Tests
Testing a robotic vehicle is nothing like taking a dealership's car out for a spin around the block. "A lot of the behaviors are hard to test," says Wicks. "You don't send a vehicle out into the streets of Blacksburg autonomously, obviously, for liability reasons and common sense."

Real-world testing barriers inspired the Virginia Tech team to build its own autonomous vehicle driving simulator. "It's a first class simulator that allows us to take the software and see how it behaves, not only with the sensors that we've modeled but also with other software," says Wicks.

Still, while simulation is useful, nothing beats real-world testing. Competing in the Urban Challenge will require researchers to work in remote, inhospitable locations for long periods. For a non-urban DARPA competition held in 2005, Montemerlo and his team lived in the desert, full time, for two-and-a-half months. "Day after day you would go out, drive the vehicle, and discover some problems," he says. "In the evening you would fix the source code and then go back the next day and do it again."

Wicks feels that readying a vehicle for a DARPA competition is a lot like preparing for any major race. "Like any competitive team, you have to practice and you have to learn to get everything to work together," he says.

Finish Line
The Urban Challenge is more a test of functionality than of speed. A vehicle that runs the course fastest won't win the competition if it consistently fails the autonomous driving tasks. And while the top teams will bask in glory, the Urban Challenge's ultimate winner will be the military, which will be able to cherry-pick the best autonomous technologies. Whitaker expects the military to quickly roll out a multitude of autonomous vehicles. "In the short run, you'll see autonomous snowplows on airstrips and vehicles in controlled environments, like in ports or on military bases, being able to operate autonomously," he says.

Commercial automakers also stand to benefit from the Urban Challenge, since they're sponsoring technologies such as automated parking and collision avoidance that will be incorporated into their product lines. In fact, the Urban Challenge provides a nearly ideal test bed for cutting-edge vehicle operability designs, says Varsha Sadekar, group manager of the active safety and driver assistance team at the General Motors Research and Development Center. GM sponsors Carnegie Mellon's Urban Challenge entry, a robotized 2007 Chevrolet Tahoe.

Meet The Boss
Carnegie Mellon University's DARPA Urban Challenge entry, "Boss," scrambles across the dusty, hardscrabble landscape at the General Motors Desert Proving Grounds in Mesa, Ariz. The driverless SUV, one of two contributed to the school's project by GM, carries a full load of autonomous driving technologies, including automated throttle, braking, and steering control functions.

"It's also been equipped with a lot of different sensors," says Varsha Sadekar, group manager of the active safety and driver assistance team at the General Motors Research and Development Center in Warren, Mich. "Radar and lidar can look at the short range as well as the long range," she notes. "Plus we have cameras on the vehicle and, of course, we have GPS for positioning." A central computer, supplied by Intel, runs planning software that continuously tells Boss where and how to drive, how to stay out of trouble, and how to most efficiently reach a destination.

Sadekar says it's important for automakers to invest in the future. "Autonomous vehicles clearly will not happen overnight, but it gives us a nice thing to aim for," she says. "In the meantime, we get experience on developing the features that we will be able to give to our customers as we move toward that big goal: the self-driving car or truck."

Drive And Park
Besides hoping to win the competition, the team, headed by robotics professor William "Red" Whittaker, is looking to develop autonomous driving technologies that can be used in production vehicles. Self-parking is one key area the team is targeting.

As any driver knows, parking is a talent that can take a human many years to perfect—some people never quite manage to master the skill. For Boss, parking is largely a matter of deduction. The vehicle must flawlessly execute a series of decisions, often involving forward and reverse motions, and perform precise tight turns to position itself correctly (Fig. 3). So far, Boss has contended with parking lots cluttered with obstacles, yet still needs to improve its ability to navigate around moving obstacles, such as lighting pylons and shopping carts.

Hands Free
Sadekar says it's only a matter of time before autonomous driving becomes a mainstream technology. "I can see, within the decade, people having autonomous vehicles that would drive on freeways," she says. In fact, one autonomous function, adaptive cruise control, has been offered with some GM vehicles since 2004.

"With traditional cruise control, you can set a speed and your car will keep going at that speed," Sadekar says. "With adaptive cruise control, if someone cuts in front of you, your car automatically slows down to a point where it keeps a certain distance between you and the car ahead of you."

GM is now looking to combine adaptive cruise control with "lane keeping," a function being tested on Boss. "With lane keeping, cameras look at lane markers on freeways and can automatically steer the vehicle to keep you within the lane," Sadekar says. "So now, you have two of these elements where you can keep yourself in the lane and you can keep a safe distance from the guy in front of you."

With Boss, the Carnegie Mellon team hopes to live up to the impressive track record set by the vehicle's namesake, Charles F. "Boss" Kettering (1876-1958). Founder of Delco and GM Research, Kettering was a prolific inventor whose creations included the all-electric ignition system, the lightweight diesel engine, and safety glass. "Boss is something that would have really excited Kettering," Sadekar says. "It's a way of keeping his spirit alive."

TAGS: Robotics
Hide comments

Comments

  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.
Publish