By making cars that don’t crash, you can take the mass from 3,000 or 4,000 pounds down to below 1,000 pounds. And if you’re designing them for major cities where the average speed is less than 20 miles an hour, you can say that a top speed of 35 miles an hour may be sufficient.
Once you’ve determined that a car can weigh less than 1,000 pounds, that it needs to have a top speed of only 35 miles per hour, and that it doesn’t crash, you start looking at how many parts and how much mass you can take out. And lo and behold, you find that lower mass begets lower mass. You find yourself chasing it down, and start thinking about the minimum machine that can meet the requirements of moving around and interacting in cities in a better way than we have today.
S+B: Other than safety, were there certain key developments and ideas that influenced your thinking?
BURNS: There were several developments that were transformational. One was the development of OnStar, GM’s on-board telematics system that uses GPS. When it was first developed in the mid-1990s, we thought the main applications would be tracking the car if it were stolen, deploying the airbags and sending an emergency signal in the event of a crash, or unlocking the doors if you locked your keys in the car. The technology to include the kinds of navigation systems that are common now didn’t exist. But as the technology evolved, it enabled the increasingly sophisticated GPS navigation systems that have become so enormously popular.
But the real insight was when we realized that once you have two vehicles with sophisticated GPS systems like that, you can determine their proximity to each other within 1 meter, and if you then share the data from their stability control systems, you can predict where each will be within the next 20 milliseconds. And once you can do that, you can program the cars’ systems to prevent them from running into each other.
Another transformational moment was the DARPA [the Defense Advanced Research Projects Agency] Urban Challenge in 2007. It was a contest in which teams had to enter a driverless car that could complete a 60-mile course in less than six hours, obeying traffic rules and safely avoiding other traffic and obstacles, held at a military base in Southern California. We formed a team with Carnegie Mellon University, and we won the competition. Out of 85 vehicles entered, three others finished the race. That showed that you could design a car that could drive itself — and it was done without using GPS for navigation. Once you add GPS and the idea of the Mobility Internet, it becomes much less complicated.
Another series of insights came from the development work we were doing on new concept cars. In the late 1990s, I worked with Byron McCormick, who headed our fuel cell program but was also lead developer of GM’s stability control system. He knew a lot about the four corners of the car — that little patch of traction, torque, steering, chassis dynamics, and braking where the rubber meets the road. He got me passionately interested in the idea of using software to control what’s happening at that point of interface with the road, and what electric drive would enable in terms of changing the driving dynamics of the car.
At about the same time, Chris Borroni-Bird joined GM, and we teamed up to build a concept car called the Autonomy, based on drive-by-wire designs like those used on airplanes, controlling electric motors at the wheels. It had a skateboard-like chassis, with electric wheel motors powered by a fuel cell, with a body that docked onto it the way a laptop fits into a docking station. I had the honor of unveiling it at the 2002 North American International Auto Show, and there was a collective gasp from the audience. It was inspiring to realize that people were looking for something like that, and that no one had really visualized for the world what you might be able to do with a whole new design palette and what it could mean for the future of the automobile. In 2005 we built the Sequel, a car that embodied all those ideas, which we drove 350 miles from Rochester, N.Y., to New York City on a single tank of hydrogen.