The car that drives itself

(Courtesy of Marcial Hernandez/VW)

In 1984, Michele Mouton was the first woman to win the Pikes Peak International Hill Climb–a twelve-and-a-half mile race to the Colorado mountain’s summit. About 26 years later, professor of mechanical engineering Chris Gerdes and his Dynamic Design Lab are preparing to run Mouton’s namesake, a white Audi TTS named “Shelley,” up the mountain again.

But this time, there won’t be a driver.

Gerdes’ team of graduate and Ph.D. students and Volkswagen’s Palo Alto research lab have spent the last two years conceptualizing and modifying the car to make the solo climb up the Peak.

At this point, Shelley knows enough to navigate parking lots and racetracks around Stanford. Rami Hindiyeh, a second-year doctoral student in mechanical engineering, said, “the basics are working pretty well.”

But not all systems are “go” yet, and the car’s computerized brain is still being refined. At a recent driverless speed test in Bonneville Salt Flats, Shelley ran into some problems when she passed 130 miles per hour.

“Our GPS started giving us some issues once we started going that fast,” Gerdes said.

Gerdes’ team’s work is a variation on one theme: make Shelley drive faster, safer and smarter.

“The one thing that we’re trying to emphasize with the project is it’s extremely fun to build a robotic racecar, but the goal is not to have robotic car races replace racecar drivers,” Gerdes said. “The real belief is that if we can learn how to control a car at its very limits of handling, then we can also help ordinary drivers who enter a turn too quickly or are driving on a wet road and don’t realize when they need to brake…so that’s ultimately where we hope this goes, is safety systems.”

Shelley isn’t the first of Stanford’s robot cars–before the Audi TTS, there was Junior, Stanford’s autonomous Volkswagen Passat.

“Junior was a perceptual challenge,” Gerdes said. Junior, and its predecessor Stanley, were designed to perceive the environments around them, understand signs and recognize the “driving situation” of nearby vehicles<\p>–<\p>then logically respond to what they saw.

But Stanley and Junior crept along at speeds well below the average driver’s comfort level, and there was little emphasis on driving dynamics.

“They’re both autonomous vehicles to some extent, but they both have very different scopes, and I guess you could say, very different personalities as well,” Gerdes said.

Gerdes’ group wanted to avoid doing what they had done before, and wanted to push technology forward in other areas.

“Can we go around turns as fast as possible, brake at the last possible minute, can we accelerate out as soon as they’re steering out of a turn?” Gerdes said, listing what the team is working to perfect in Shelley.

Hindiyeh has the task of crafting Shelley’s “judgment.” He writes software designed to mimic a rally car driver’s mind with a series of mathematical analyses that predict how the car should control itself in different situations. Currently, he looks at “ways to slide Shelley through turns like a rally car racer would.”

Mick Kritayakirana, a third-year graduate student in mechanical engineering, is in charge of working to control Shelley “at the limits, kind of like racecar drivers race on the pavement,” he said. He is working on the autonomous racing controller.

The standard Audi TTS’s steering, brakes, gears and throttle are all controlled electronically, so Shelley required few mechanical modifications by Gerdes’ research team to integrate her systems into a “controller area network,” that allows the vehicle’s components to communicate. The network allows the team to individually switch each component from manual to automatic so they can test its reliability.

Shelley’s most critical components are a rack of GPS antennas coupled to a system that determines speed and sideways motion. It controls the car’s direction when the GPS can’t connect with the satellites, and gives prompt updates on the car’s position.

Yet while the combined effects of Shelley’s systems are complex, the computer in the car’s trunk that processes these data isn’t any faster than one you could buy a decade ago. Most calculations are done separately within the GPS and in the vehicle electronics.

“We don’t need a whole lot of computational power to run the driving and racing algorithms,” Gerdes said.

Testing Shelley today usually means having a safety driver in the cockpit, in case something goes awry with one of its electronic systems. In the near future, however, the team hopes Shelley will be able to flawlessly navigate solo.

“We have to spend a lot of time trying to make the car listen to what we command,” Kritayakirana said.

The Pikes Peak course will be plotted on a GPS map for the car to follow, and based on that information and how much friction the computer predicts, it has an idea of how fast it can take turns at different angles and with varying road surfaces. The computer refines its speed and steering with each test turn to figure out what Gerdes calls Shelley’s “braking point.”

David Hoffert, a second-year graduate student in mechanical engineering, explained that Shelley tries to model what a human being would do and plan ahead, with a safety net so that “when we screw up, [it makes] the appropriate correction.”

“When a human is driving a car and they see a turn coming up, they can just try to, at a constant rate, so to speak, just try to turn the wheel towards that curve preemptively,” Hoffert said. “And that works because roads are designed with certain mathematical geometric properties that if you do that, [you] follow the path.”

As Gerdes’ team nears the finish line, they are continuing to closely collaborate with Volkswagen’s research group. They have “weekly meetings where we talk about our current status…and evaluate the hardware and software,” said Marcial Hernandez, senior research engineer at Volkswagen.

Gerdes’ team aims to have Shelley on the mountain in September, right before the start of fall quarter.

“We’d really like to send the car pretty close to its capability, certainly much, much faster than people would be comfortable driving unless they were highly skilled racecar drivers,” Gerdes said.