Читать книгу Autonomy: The Quest to Build the Driverless Car - And How It Will Reshape Our World - Lawrence Burns - Страница 9

An engineer is someone who likes to work with numbers but doesn’t have the personality to be an accountant.

—UNKNOWN

Over the last fifteen years of development on autonomous vehicles, if there is one figure who has been there, on the ground, getting his palms filthy with engine grease, breathing carbon monoxide exhaust and burning himself with electronic solder to solve each little problem as it comes up, it is Chris Urmson. The technical lead of the Carnegie Mellon University teams that competed in the three robot-car challenges staged by DARPA, Urmson’s also the figure anointed as leader by the founder of Google’s Chauffeur self-driving car project, Sebastian Thrun. In fact, Urmson ran day-to-day operations on the team from its founding in 2009 to shortly before the spinning out of Chauffeur from Alphabet into a stand-alone company, known as Waymo, in 2016. Finally, Urmson also played a key role in the power struggle that dominated Chauffeur for long periods of its existence.

To make this thing work, Urmson has sweated blood.

He’d be the first to admit he doesn’t have the outright incandescent charisma of some of the other figures in this story. Urmson is smart, sure. He refined his willingness to consider every possible solution to a problem, no matter how outlandish, in the creative-thinking challenges that dominated the Canadian educational system’s classes for gifted learners. What Urmson lacks is the bumblebee attention span of some of his self-driving colleagues. Perhaps this is because of the milieu in which he was raised. The oldest son of a prison warden and his nurse wife, Urmson grew up in small Canadian cities—Trenton, in eastern Ontario, where the biggest employer is a military base. Victoria, the seat of the British Columbia provincial government. The not-exactly-bustling metropolis of Winnipeg, Manitoba. His dad was rising through the ranks of the northern nation’s correctional services bureaucracy, eventually running not just one prison, but a whole area of them, until the family settled in the sleepiest city of them all—Saskatoon, the capital of Saskatchewan, the least assuming province in one of the least assuming countries of the world.

Urmson grew up among people who viewed with suspicion those who drew attention to themselves. What the guy is, is solid. Straight-shooting. Steady. Urmson is not the guy you’re going to notice first when you walk into a room. But you spend enough time with the people in that room, and I don’t care who is in there, after a while Urmson will be the guy you trust to lead—to carry out the plan.

And in April 2003, Chris Urmson had a plan. In fact, Urmson thought he had the next couple of years of his life pretty well figured out as he drove from the remote Chilean city of Iquique to the great salt plains of the Atacama Desert. The road from Iquique into the Atacama would make anyone nervous. It zigs and zags from the Pacific Ocean on up a near-vertical shelf. Those who remember plate tectonics from high school geology might recall that this is where the Pacific’s Nazca Plate collides with South America, pushing the continent into the air, creating a ridge thousands of feet high and a rain shadow that runs six hundred miles up and down the Chilean coast. That rain shadow is the Atacama. One of Earth’s most forbidding landscapes, the driest nonpolar desert on the planet, an area so desolate that scientists use it as a stand-in for Mars. That was what Urmson was doing there. He was one of a handful of roboticists joining a team of NASA staff members to test a robot designed to crawl across the Martian landscape to seek out signs of life.

At twenty-seven, Urmson was tall and athletically built, with sandy blond hair and smiling blue eyes behind round, wire-framed spectacles. He tended to jam a baseball hat so low over his brow that the brim would touch the top of his glasses. Urmson planned to spend about a month in the Atacama. Then he’d return to Pittsburgh, where he was a graduate student in the robotics program at Carnegie Mellon University. He’d write up his dissertation, undergo the grilling every thesis committee is supposed to do, hopefully get his PhD and then a job—maybe join the faculty of his alma mater’s Robotics Institute, home to more robotics brainpower than anywhere on earth, or maybe join one of the start-ups that occasionally spun out of the university. In any event, he’d start making money, enough for him and his wife to have the kids they’d been putting off while Urmson finished his studies.

The campsite that Urmson’s research group chose amounted to little more than a handful of bright yellow dome tents, a slightly bigger meeting tent—where they kept the computers—and a pickup. And Hyperion. Hyperion was the robot. Not the conventional kind of robot. No arms and legs. Rather, Hyperion sat on a quartet of bike tires and was roofed with solar panels and powered with an electric motor. Hyperion was the reason why Urmson, and his fellow scientists from Carnegie Mellon and NASA’s Ames Research Center, had traveled over half the planet.

Hyperion was designed to wander across the Martian surface, sniffing and scraping and testing the soil for signs of life. Urmson was in charge of programming the software that dictated how fast Hyperion rolled.

The scientists took their breakfasts and dinners at a nearby salt mine. Nights, they sat around a fire and watched the camanchacas roll in, the Pacific salt fog that could rust exposed metal within a single night. They turned in to tents they used for warmth rather than protection. You camp in other deserts, you need a tent to keep snakes out of your sleeping bag and scorpions out of your boots. But nothing lived in the Atacama Desert. Not snakes. Not scorpions. The only living things that Hyperion’s minders saw were vultures.

The encounter that would change Urmson’s life started with the sight of a long dust cloud led by a speeding pickup. Some minutes later, the dust cloud followed the pickup into the Hyperion campsite. The door opened, and out of the truck popped William L. Whittaker, commonly referred to as Red.

Whittaker was another big guy, an inch or two taller than Urmson at about six-foot-three, with shoulders that look like they’d brush the sides of interior door openings. His scalp is closely shorn; years ago, when he did have hair, the color of it was what gave him his nickname. His gaze is intelligent and contemplative. It feels like his eyes can see right into your soul when he looks at you. Anyone who spends five minutes with Whittaker can tell that he spent formative years in the U.S. Marine Corps. He speaks in the sort of aphorisms that drill sergeants put on their bedroom walls. “Winning isn’t everything,” he might say. “It’s the only thing.” And: “Worry is a formula for failure.” Another favorite: “If you haven’t done everything, you haven’t done a thing.” Hyperion was somewhere around the sixty-fifth robot Whittaker had worked on in his career as a roboticist.

The Carnegie Mellon professor strode out of the pickup in boots, conducting a round of handshakes with his big hands. He was there in part because he was Urmson’s thesis adviser, and he was checking on his charge. But you could tell that Whittaker was holding something in. Something big. Pretty soon, Whittaker came out with it. The U.S. Department of Defense was staging a driving race for robots. Specifically, the Defense Advanced Research Projects Agency. DARPA, Urmson knew, was the U.S. government’s developmental laboratory, credited with spurring such useful inventions as drone technology and the Internet (a military invention whose distributed knowledge network was intended to safeguard the records of the U.S. government in the event of a nuclear attack). DARPA was also responsible for such less-than-useful innovations as mechanical lobsters for the U.S. Navy and DNA-editing techniques intended to create humans who didn’t need sleep. Now DARPA Director Tony Tether was turning the agency’s direction toward autonomous cars.

For years, Washington had pushed American defense contractors to develop autonomous technology so that a third of all American military vehicles could be self-driving by 2015—a stated mandate from Congress. In the aftermath of 9/11, the effort took on added urgency as the U.S. military lost infantrymen and -women to improvised explosive devices planted under the roads in Afghanistan and Iraq. If self-driving vehicles ever became possible, military robots might drive themselves over the sort of desert roads found in overseas theaters of war. But the four-star generals had been frustrated with the pace of change. The problem was proving too difficult for the military contractors. And so Tether struck upon a novel solution: DARPA would stage a race. For robot cars.

As Whittaker recounted the details to Urmson, they sounded a little insane. DARPA said it would allow any American team to enter—student, hobbyist, professional, whomever. The course would bisect the Mojave Desert, running eastward from Barstow, California, to Primm, Nevada, for a distance of about 150 miles. The prize money would go to the first team that could do it in under ten hours.

“Wow,” Urmson said, thinking Whittaker was just making conversation.

But Whittaker never just made conversation. The prize money, the old marine said, was a million bucks. And Whittaker wanted to win that money with Urmson’s help.

It would be three years before I met Chris Urmson, who would go on to become one of my favorite people. But I can see how this situation would have presented him with a dilemma that contradicted two of his prime directives. Urmson had a seemingly innate desire to try to improve the stupid and inefficient things about the world; he once interrupted an important business meeting at a Pittsburgh coffee shop to burst out onto the street and direct traffic, just to help someone turn left out of a parking lot. He was programmed with an engineer’s duty to seek out the coolest and most interesting projects that could change the lives of the most people. Which is why Hyperion was such a perfect project for him. How could you get cooler than an autonomous robot designed to seek out life on other planets?

Actually, it turned out that you could. Urmson’s work with Hyperion was helping the robot travel anywhere from 15 to 25 centimeters a second—about the pace of a slow walk. In the DARPA race, the robot would have to travel 150 miles in at least 10 hours, which required an average speed of about 15 mph, as fast as most cyclists went. The speed, the money, the fact that the race was intended to address an issue killing American soldiers overseas—Urmson got it. He ached to participate.

But there was a problem: He was also programmed with a duty passed down to him from his parents, to do what was best for his family.

Chris Urmson was born in 1976 to Paul and Susan Urmson, an English couple who had immigrated to Canada because they thought it would represent better opportunities for their three sons. Paul’s first career was as an electrician, and then, once his kids were born, he pursued his college degree at night school, earning his BA and then master’s. Susan enrolled in nursing school after the kids were born and went on to administer methadone programs within the Canadian prison system.

The point? The three Urmson boys grew up in homes where the parents were always working, always bettering themselves for the sake of the family and where education was prized from the kids’ earliest ages. The Urmson parents ran their lives for their children. The family moved a lot because Paul’s work in the prison system required him to transfer around the country. Each time they did, Paul and Susan settled the family in the cheapest house in the nicest neighborhood they could find—a strategy they devised to send their kids to the best public schools. The strategy worked. In addition to birthing one of the most important engineers in the development of autonomous cars, the Urmsons also raised an orthopedic surgeon and a Mountie, a member of the Royal Canadian Mounted Police, which is something of a trifecta for middle-class families north of the border.

At a young age, Chris’s teachers assessed him as gifted, which qualified him to attend special classes with similarly intelligent children. The classes provided the ability to conduct independent projects. Gifted-program teachers encouraged their students to enter a series of science fairs then known as Olympics of the Mind, which challenged participants to solve unconventional problems. How do you build a tower out of just paper towel tubes? Propel a toy car with a mousetrap? Safeguard an egg dropped from an extreme height?

The experiences set Urmson up well to compete in Canada-wide science fairs. The year the Urmson family moved from Victoria to Trenton, the national finals happened to be held in Victoria. Urmson ached to visit his old friends, and so he directed all his energies toward winning the local competition. His entry, “Striking News About Impacts,” predicted the direction a body would travel after a collision. He won the Trenton fair, and received the free trip to Victoria.

Bit by the science bug, Urmson followed up with a project involving a model of ionic propulsion—“Ionic, Isn’t It?” was the project name. It not only won him another trip to the Canada-wide competition, but also garnered him second prize. Another year he won a silver medal at the national level and qualified for a four-week trip to study programming at Israel’s Weizmann Institute. Urmson would go on to study computer engineering at the University of Manitoba, where one of his projects entailed building a robot that traveled autonomously around a darkened room, seeking out the brightest sources of light.

Urmson was torn in his last year of university. One path, favored by mothers everywhere, might have seen Urmson going on to med school. Except that didn’t exercise his yen for building things, for envisioning complex systems and then figuring out how to make them work. Wandering by the office of his computer engineering department one day, Urmson’s eye was caught by a remarkable poster: a vehicle, maybe some sort of a planetary rover, climbing up and out of some sort of crater. “Come be a part of the robot revolution!” the poster read, with information about attending Carnegie Mellon University. It was a career based on the sort of thing Urmson had been doing all his life. Olympics of the Mind. Science-fair stuff. He applied, and ended up in Pittsburgh the following year.

At Carnegie Mellon, Urmson met Red Whittaker, who by 2003 already was a legend in American robotics and one of the best-known robot designers in the world. Born in 1948, Whittaker was fifty-five in 2003 and had become widely known for his willingness to take on projects that everyone else thought impossible. “If there is anyone in the world who can find a way to make things happen, it’s Red Whittaker,” said one colleague.

Whittaker may have been genetically programmed to ignore the impossible. His father was an air force bombardier in World War II who would go on to sell explosives to mining companies. His science-teacher mother was an amateur pilot who once flew under a bridge while the young Whittaker was in the plane alongside her. After serving in the marines for two years, Whittaker attended Princeton University, earning his degree in civil engineering in 1973, and then attending graduate school at Carnegie Mellon.

Whittaker made his name after the partial meltdown in 1979 of the Three Mile Island nuclear-generating station, America’s worst-ever nuclear accident. Cleaning up the incident required getting into the reactor’s basement to learn how radioactive the site was. Several contractors spent almost a billion dollars on the cleanup but still couldn’t figure out how to get inside. When Whittaker asked for his shot, the government figured they didn’t have anything to lose. Whittaker reasoned that, while the radioactivity prevented humans from getting into the reactor, machines should have no problem. He created a three-wheeled Remote Reconnaissance Vehicle, known as “Rover,” which he operated by remote control. Rover successfully made it to the basement. Best of all? The program cost only $1.5 million, which the government considered cheap.

Since then, Whittaker had specialized in building robots designed to work in harsh environments. One of his creations explored volcanic craters. Another mantis-like contraption built structures in space. Still another, created with a team that included the German software wizard Sebastian Thrun, crawled through the darkness of long-abandoned mines, mapping their interior passageways. Urmson had worked with Whittaker to develop computer algorithms designed to increase the speed at which robots were able to travel autonomously.

When Urmson returned from the Atacama Desert, he had a tough conversation with his wife, Jennifer. Urmson wanted to set aside the completion of his PhD for a time and pursue the DARPA race with Whittaker. The DARPA Grand Challenge was the talk of their academic specialty. DARPA had figured it might be lucky to get twenty entries. Eventually, 106 teams would enter. Urmson felt he had no choice but to join. Who knew what sort of fascinating epiphanies would emerge from the project? Who knew what Urmson would miss if he didn’t take part?

Urmson convinced Jennifer to let him do one race. The couple would put off having kids until Urmson was done. But then fate threw them a slider: It turned out Jennifer was already pregnant. The news added to the pressure Urmson felt to win. After all, it was the best way to ensure he’d get a high-paying job, once he finished.

To attract a team, Red Whittaker put up posters all around the Carnegie Mellon campus advertising an unconventional, graduate-level seminar class, Mobile Robot Development. It was pass or fail, and it featured only one assignment: to build a robot that would win the first DARPA Grand Challenge. He also sent out email blasts to potential sponsors and volunteers, which featured his trademark bravado: “The race defies prevailing technology, and many hold that the challenge prize is unwinnable in our time.”

Whittaker staged the first meeting of the team in a Carnegie Mellon seminar room on April 30, 2003, according to Wayt Gibbs, a reporter the magazine Scientific American embedded in Pittsburgh. “Welcome to the first meeting of the Red Team,” Whittaker began. “I am committed to leading this team to victory in Las Vegas next year.”

The men and women in the room were about as motley a crew as it was possible to put together in Pittsburgh tech circles. Bob Bittner was a former combat engineer who’d spent six years in a submarine. Spencer Spiker was a retired helicopter test pilot, a West Point–educated mechanical engineer who led two hundred people as a company commander in the U.S. Army, and who had left the service to spend more time with his family—then found himself jobless in a severe recession. He joined Red’s team because he had nothing better to do, then worked himself into a full-time staff position. Michael Clark was a NASA engineer who used a wheelchair to get around; having fallen on hard times, he lived for a spell out of his van. Lots of people had seen Red’s poster, apparently, and lots of people were inspired to work on the project it advertised. “I don’t know anything about computers—but I’d like to volunteer,” said Mickey Struthers, a postman who showed up to the first class because he wanted to participate in a historic science project.

“You’ve got a warm body”—Whittaker grinned, shaking Mickey’s hand—“and we need warm bodies.”

They began their efforts brainstorming what sort of a vehicle they would use. DARPA had announced the course would be designed by Sal Fish, the operator of such tough off-road races as the Baja 1000. Red Team figured it would have to be prepared for a course that would wind through dry river gulches, box canyons, mountain ridges, rocks, sagebrush and cliffs. So the robot they designed had to be able to either get around such land features or drive over them.

No idea was too outrageous to be considered. One of the first suggestions was a giant tricycle that had wheels seven feet in diameter. The team discussed using a Chenoweth combat dune buggy, a low-slung contraption on four fat tires favored by mercenaries and warlords. Other brainstorming options included construction equipment, an all-terrain vehicle and a tank. But the team ultimately opted for pragmatism. After all, Whittaker figured the budget to develop the robot would be around $3.5 million. Labor aside, $725,000 of that entailed the cost of the products required to build the vehicle. Whittaker was crisscrossing the country to find sponsors. Intel, Boeing and Caterpillar all kicked in some money. Google, which everyone thought of as a search engine company at this point, sponsored Red Team to the tune of $100,000 after Whittaker visited its headquarters in Mountain View, California, and met both Larry Page and Sergey Brin. But such funds wouldn’t go far when you were trying to build the fastest-ever robot car. Red had bought a cattle ranch a couple of hours east of Pittsburgh in the early nineties because he felt his academic life was too sedentary and sought physical activity that worked his muscles, not his mind. In September 2003, with the March 2004 race date fast approaching, Whittaker finally bought the vehicle that would become their robot from another farmer in the area.

Some in the class were astonished when they saw it. Shouldn’t a self-driving car look cool, and polished and, um, high-tech? The vehicle Red had procured was the opposite of high-tech. It was a High Mobility Multipurpose Wheeled Vehicle M998: a Humvee, battered by time. It was seventeen years old. No one had any idea how many miles it had, because the vehicle didn’t have an odometer. Nevertheless, the price was right: $18,000. The key thing was, it worked.

Whittaker was under a lot of pressure. Around the country, dozens of robotics enthusiasts were working to create entries for the challenge; so many, in fact, that DARPA was requiring everyone to submit a detailed and academically rigorous declaration of the approach they planned to take. The step was intended to limit the race entrants to serious competitors. There were high school students and bored mid-career engineers. Several were former contestants on the mechanical-gladiator game show, BattleBots, which featured remote-controlled robots fighting to the death, or at least, deactivation. Regardless of where they came from, the competitors all seemed to have one goal in mind: Beat Red Whittaker’s team. Why was the CMU team in so many other people’s sights? Whittaker’s team was the biggest, with thirty members. It was one of the best funded. And many also believed that it was DARPA’s hoped-for winner.

Red’s leadership style was to take a bunch of people, introduce the problem to them, set ambitious and clearly defined goals that reflected progress toward the solution—and then get out of the way. He’d drop in regularly to check in and apply pressure on his charges. Such visits could be intense. According to a Wired article, Whittaker once drew an analogy between developing robots and the labor required to construct the enormous historic monuments around the Nile. “If you’re in Egypt building the pyramids, you’ve got to have slaves,” he said. The implication? Whittaker’s students were his slaves. One of Red’s longtime students, Kevin Peterson, who would become Red Team’s software lead, had attended Princeton High School, where he encountered Dr. Anthony Biancosino, the domineering music teacher on whom Damien Chazelle loosely based the bandleader in the 2015 movie Whiplash. Peterson responded to Whittaker’s style, he says, in part because he’d already been through his experiences at Princeton with “Dr. B.” “There was an ethos around both of them of being larger than life and somewhat mysterious,” Peterson recalled. “The idea that you need to work hard to be part of their exclusive team if you want to join them. They’re both up to big things and you need to be a badass to be on the team. Funny thing is, both of them would accept and build anyone who had that level of dedication. It’s more about hard work than initial skill.” One of Whittaker’s favorite motivational anecdotes placed his charges in the roles of the Inuit in the Arctic, who had to decide which strategy to use when seeking out food. Are you going to go out and try to find a few berries and bits of lichen? Whittaker would ask his team. Or are you going to find and kill the walrus that feeds the whole village?

Sometimes it was hard to tell what Whittaker meant by his stories. Peterson interpreted this one as a challenge. Were you going to go about your life just getting by? Or were you the type who was going to go out and give your best effort to do something awesome?

Realizing that his course would require more work than they were prepared to give, some people dropped Whittaker’s class. The ones who remained essentially dropped every other one of their classes and just worked for him. Peterson was one of the ones who remained. He gave up his social life, as well as communicating with his family. He even gave up sleeping. Several months in, he became so sleep-deprived that he fainted. The problem was that he was going down a set of stairs when he did. He hit his head, was taken to the hospital to be assessed—and was back working on the project within a few days.

Empowering inexperienced and sleep-deprived graduate students who were totally committed to the project’s success could create some unusual situations. One morning, Whittaker and Urmson arrived to check in on the students and volunteers and were met with the results of one of these hyper-caffeinated work sessions: Their treasured Humvee no longer had a roof. Working through the night, one of the student team members had decided that the Humvee’s interior didn’t have enough room to store the batteries and computers and actuators that the self-driving equipment would require. So he went and got a Sawzall and cut through each one of the Humvee’s roof pillars, essentially decapitating the vehicle.

This was the sort of initiative that would typically have been applauded by Whittaker. Except the impromptu roof amputation wasn’t really necessary. Even if the equipment couldn’t fit in the Humvee’s cab, they could have ripped out some seats, or mounted additional equipment on the Humvee’s roof. Removing it made the vehicle illegal to drive on public roads. From then on, whenever they wanted to take the Humvee to the sort of wide-open space where they could test it, they would have to tow the vehicle—an ignoble start for a robot that was supposed to drive itself.

To provide the Humvee with the ability to drive itself, Red Team essentially reverse-engineered the sensory tools humans use to help them drive. The vehicle needed, for example, eyes to see—and so the Red Team procured several types of LIDAR (Light Detection and Ranging) devices. The LIDAR’s job was to shoot out beams of light and sense when the beams bounced back. Precisely calculating the timing of the beams’ return allowed the LIDAR to determine how close the sensor was to the object that the light beam bounced against. Repeated thousands of times per second, the LIDAR could create a rudimentary picture of the world outside the vehicle.

The main LIDAR sensor would allow the robot to detect obstacles seventy-five meters ahead. Three supplemental LIDAR devices scanned a wider field of view within twenty-five meters of the robot’s front end. A stereo-vision processing system represented a different way to use light to detect objects, employing a pair of cameras. But the cameras and LIDAR might have trouble penetrating the dust clouds that could arise on sandy desert roads. To provide a sense of the world in dusty conditions, Red Team also bought a radar system that used sound to detect obstacles.

To control the vehicle’s direction and speed, Red Team wouldn’t be able to use a foot on the gas pedal or a hand on the steering wheel. Actuators would take their place. Essentially, these were electric motors that twisted, pushed or pulled—to make the vehicle accelerate, brake or turn left or right.

Sitting in the center of all that was a series of computers, the robot’s brain. Donated by Intel, one was a quad-processor Itanium 2 server that featured three gigabytes of RAM. Some of the computers were intended to combine the information provided by the LIDAR, the stereo-vision system and the radar sensor to create a model of the world. Another computer employed GPS data and motion-tracking tools to locate the robot in the world within a single meter of accuracy. Now that it had a conception of its surroundings and knew its location, the robot’s computer system would have just two questions to answer. Two questions that humans asked themselves, thousands of times a trip: How fast should I be going? And where should I be steering?

Whittaker scheduled one hundred days to actually get the robot assembled and the software built. The deadline fell in November, but as Thanksgiving approached, significant portions of the vehicle remained unfinished. The computers weren’t wired together, for example. Nor were the sensors mounted. The robot did have a name, though: Sandstorm, after the dust clouds the vehicle would kick up in the Mojave.

Whittaker and Urmson both worried a lot about the Mojave Desert. They worried about the off-road conditions of the course, and the effects of the Mojave’s rutted roads on their sensitive sensors and microprocessors. Driven over even at moderate speed, the Mojave Desert’s rocks and ridges were bound to create vibrations that the students believed had the potential to damage the computer’s memory. After all, your basic disk drive is just a magnetic metal plate that spins really quickly. They’re encoded by a precise bit of metal that hovers just above the plate. Extreme shocks could see the metal stick gouging chunks from the spinning plate and damaging the drive. Those same bumps could create false readings from the sensors.

Consequently, Red Team spent a lot of time determining how to insulate the computers and sensors from the jars and bumps that would happen as the Humvee drove across the desert. The solution, they decided, was to protect the equipment the same way automobile manufacturers insulated humans from bumps and jars. With springs and shock absorbers, which were fitted to an enormous metal box where the Humvee’s roof used to be. Dubbed the “e-box,” for electronics box, the 1,200-pound container didn’t just contain hard drives. It also encompassed much of the robot’s most sensitive equipment—the computers, the GPS system, the radar as well as the supplementary LIDAR units.

The main LIDAR and the stereo-vision device still remained sensitive to the pitches and rolls that could strike the robot as it navigated the off-road trail. So the team spent untold hours engineering a device based on old nautical gimbals, complex series of interconnected arms and pivots that kept a ship’s compass stable in even the heaviest of seas. Part of the Red Team designed and built their own gimbal, mounting inside it the main LIDAR and the stereo-vision system, and protecting it all in a sphere a little larger than a classroom globe. Little motors in the gimbal allowed Sandstorm to direct the LIDAR and camera wherever it needed to sense the world. Heading into what its onboard map told it was a leftward curve, the LIDAR would “look” to the left so that it could see the world in the direction of the world to come.

As technical director, Urmson was the one in charge of putting all these pieces together. He felt enormous pressure both at home and on the Red Team. That September, his wife had just had the couple’s first child, a baby boy. But Urmson couldn’t spend much time at home. He had made a promise to Whittaker that the robot would drive itself the entire length of the race, 150 miles, by midnight on December 10, 2003—three months before race date.

To meet that deadline he was working sixteen-hour days, seven days a week; during one furious round of assembly Urmson didn’t sleep for forty hours. The week before Thanksgiving, Whittaker added to the pressure. “This vehicle hasn’t rolled so much as a foot under its own control,” he said during one meeting with Urmson and other key team members, according to the journalist Wayt Gibbs. “You have promised to get 150 miles on that beast in two weeks … Anyone who thinks it is not appropriate for us to go for 150 miles by December 10, raise your hand.” Silence. Not a single person elevated an arm. Whittaker smiled, according to Gibbs, and made an observation in his characteristically florid language: “We’re now heading into that violent and wretched time of birthing this machine and launching it on its maiden voyage.”

The assembly work happened in a big garage in Carnegie Mellon’s Planetary Robotics building. Envision the best mechanics shop you’ve ever seen, and you’ll be close to this workspace. The ceiling is a few stories tall, with gangways and a small-scale version of a crane, the better to lift heavy objects. Lathes and drill presses, drawers full of every implement imaginable, as well as computer diagnostic equipment—every available horizontal surface features tools. It is the kind of place where you could literally make almost anything.

The venue would host Urmson and the members of his team pretty much nonstop through that Thanksgiving weekend. By the end of it, enough computers were wired together, and enough sensors mounted, that Sandstorm felt like it was coming alive. It was around this period that the team found the perfect place to test their Frankenstein’s monster. There weren’t many spots with convenient access to the CMU campus where a 5,000-pound, exhaust-snorting, diesel-gulping, oil-dripping robot could push the limits of its abilities without risking civilian fatalities. It was Mickey Struthers, the postman volunteer, who thought of the solution. One day while he was driving over Pittsburgh’s Hot Metal Bridge on the way to Carnegie Mellon, Mickey noticed the lights along the shores of the Monongahela River twinkling in the cool evening air. All except for a vast swathe of dark shoreline to the right of the bridge. Mickey knew that was industrial land that had once housed Pittsburgh’s last steel mill, the LTV Coke Works, which had closed in 1998. Since then the land had sat fallow.

Struthers suggested the site to Whittaker, who loved the idea for both its convenience as well as its industrial heritage. The 168-acre land parcel housed a railroad roundhouse and numerous outbuildings and equipment that made it seem as though it was left over from the industrial revolution, connecting the team to the same brawny spirit that had built Pittsburgh so many decades ago. With a few phone calls to the wealthy family foundations that owned the land, Whittaker arranged for the team to test there.

On the second of December, the team took the first of what would become many test runs at the Coke Works. The distressed location with its spent oil cans and rusted industrial detritus seemed appropriate for the ancient-looking Humvee, which just in general seemed to have more in common with a Jurassic-era dinosaur than one of the most innovative mechanical devices ever assembled. Snow covered the ground. The temperature was eighteen degrees. “Just like the Mojave Desert, huh?” shouted one team member, according to a Wired article. (Whittaker, meanwhile, was wandering around in a knit shirt, jeans and boots he wore without socks.) Urmson climbed aboard for the first run to manually hit the emergency stop button if the robot suddenly went crazy. The robot swerved toward a precipice when first activated, then settled and drove its course as expected. After a few uneventful laps Urmson decided, at 7:51 P.M., to see what would happen when he gave Sandstorm free rein. He clambered off the robot. The team programmed in a series of GPS waypoints that drew a dot-to-dot version of an oval. Not sure whether to breathe, the team watched the robot roll along its route for half an hour, ultimately accumulating four miles. No accidents. No incidents of any kind, in fact. They were nowhere near making their 150 miles yet, but that evening, it was difficult to deny they were progressing toward their goal.

Another week passed, and late in the evening on the tenth of December, with just a couple of hours before the midnight deadline by which Urmson and the team had promised Whittaker that Sandstorm would be able to drive 150 miles on its own, the robot was not cooperating. Bugs arose in the self-driving software every time it drove more than a few laps. Urmson and his fellow teammates had been camped out for days at the Coke Works, if you called camping sleeping in your running car with the heat on full blast. Despite daylong debugging sessions, Sandstorm remained unpredictable and occasionally suicidal—lurching into a telephone pole, catching fire, becoming suddenly unable to sense GPS signals. A calm spell saw the Humvee revolving the track, again, again, again, and then for no apparent reason, swerving off course and running itself through a chain-link fence before Urmson could activate the e-stop. Sometime later, with Sandstorm liberated from the barbed wire and the deadline approaching, Whittaker gathered Urmson and everyone else around him, according to Gibbs. Sure, the December 10 deadline approached—but even if it passed, Whittaker vowed, they’d continue their work, through tomorrow, and even the next day if necessary, until Sandstorm achieved the 150-mile goal. “We say what we’ll do, and we do what we say,” vowed Red in Scientific American.

Then it started to rain—a frigid December drizzle that soaked clothing and chilled to the bone. Sandstorm was not well protected against rain. One of the dozen or so team members still on site spread a tarp over the robot’s computer equipment. Red wasn’t around. Gibbs wrote that Urmson looked at his teammates, shivering in dripping lean-tos under blankets. He thought about the possibility of the falling moisture disabling one of their sensors, or shorting out a processor. Perhaps he also thought about his wife and baby boy back home. And he decided to send the team home.

Whittaker was livid when everyone showed up to the Coke Works the following day, Gibbs reported, comparing the team leader to “an angry coach at halftime.” He ranted about all the sacrifices they’d made to try to achieve the 150-mile goal. The shop was a mess, the robot unpainted, the website out of date—all that work went undone as everyone concentrated on getting Sandstorm in the sort of shape required to make its race run. To a roomful of people unwilling to meet his gaze, Whittaker said, “Yesterday we lost that sense deep inside of what we’re all about. What we have just been through was a dress rehearsal of race day. This is exactly what the 13th of March will be like. We’re in basic training; this is all about cranking it up a notch. Come March, we will be the machine.” Whittaker concluded his venting, Gibbs reported, by asking who was willing to work all day, every day, for the next four days, until they completed their nonstop 150-mile run. Fourteen team members in the room raised their hands. Including Urmson.

Two days later, U.S. soldiers captured Saddam Hussein in a spider hole near Tikrit, and the war in Iraq dominated headlines and the cable news channels as it never had. Every day, the news seemed to feature more casualties from IEDs in Iraq or Afghanistan—fatalities Red Team members hoped the robot vehicles might one day prevent. Then the overseas conflicts supplied Urmson with an idea.

In recent years, maps had become a crucial component of successful robotics. Maps allowed robots to locate themselves in the world much more accurately than GPS alone. A technique called simultaneous localization and mapping, abbreviated to SLAM, saw a robot scan an area with LIDAR to map the permanent landmarks—in exterior spaces, things like trees, light poles, road curbs and buildings. Then, the next time the robot traveled the same territory, it would consult its map and compare its position relative to the previous landmarks, to get an ultra-accurate idea of where it was. Problem was, Sandstorm couldn’t use this technique, because DARPA was keeping the race location secret.

Then, one day, Urmson was watching coverage of the war on one of the cable news channels. The scene will be familiar to anyone who lived through the post-9/11 period—a grainy portrait of an SUV traveling fast along a remote desert road. From somewhere in the distance, a rocket blazes into the picture, collides with the SUV and obliterates the vehicle in a blast of dust and metal.

The footage of the successful deployment of a laser-guided bomb was captured by a camera-equipped drone aircraft. The drones flew above the conflicts to provide imagery of the Iraqi and Afghan territories. Drones were searching Afghanistan for Al-Qaeda hideouts that might shelter Osama bin Laden. They were scanning Iraq for nests of Ba’athist loyalists.

If the U.S. military could use drones to obtain imagery of places so hostile and remote, Urmson thought, then such imagery would soon be available for the entire world. And perhaps, Urmson reasoned, that same type of imagery could be used to simplify the robot’s task. They weren’t able to use LIDAR to scan the race course in advance, because no one on Red Team knew where the race course was, but they did know the race went across the Mojave Desert—and maps existed of that, didn’t they? In fact, portraits of the Mojave had already been built by entities like the U.S. Geological Survey and the military.

“We realized we didn’t have to do SLAM,” Urmson recalled. “Because it was becoming clear there would be a global database [of maps] available … So why not use them?”

If Red Team members could give Sandstorm an accurate map of its surroundings before the race, they could remove a time-intensive step from the computational task. The new approach reframed the challenge. The team had assumed they were trying to build a robot that could sense the world so well, it could discern a road in the desert and navigate it safely for 150 miles. Using maps meant the robot could be told in advance where the road was, and how to drive it. The method had the potential to allow Sandstorm to travel much faster than it otherwise might.

But first, Red Team’s undergraduates, pauper grad students and volunteers would have to build the most detailed map of the Mojave Desert ever assembled. It was an enormous task, but Red Whittaker’s students were accustomed to achieving enormous tasks. A portion of the team set to procuring high-res maps of the whole of the Mojave Desert, a relatively simple matter, given Whittaker’s and Spencer Spiker’s defense contacts. Now the team set about using the maps to plot routes through the Mojave. They also dispatched two engineers, Tugrul Galatali and Josh Anhalt, to drive as many roads in the Mojave Desert as possible in a rented SUV with video cameras sticking out the windows, capturing imagery from the ground in what amounted to an early, rudimentary execution of Google’s Street View idea.

The next step saw the Carnegie Mellon mapping team comparing the footage and the map to assign each area with a value—what they called a cost. So a ridge or a cliff that would wreck Sandstorm if the robot went over it would get a cost of infinity. A smooth road or a dry, flat lake bed likely would have a cost of zero. Sandstorm’s computers then were programmed to direct the robot to drive the route with the lowest cost.

One evening, with just weeks to go before race date, the senior members of Red Team met in the loft of Carnegie Mellon’s Planetary Robotics building. “We were making some progress, trying to map every trail in that whole desert,” Urmson recalls. But at some point during this meeting in the loft, Urmson realized their work wasn’t happening quickly enough. “It became clear we weren’t going to get there,” he said. Too many different potential routes existed. By the time the race date arrived, they would have mapped out only a small portion of the possible routes.

That was the point that Red Team came to its second epiphany. To reduce the possibility of exactly this sort of advance route planning, DARPA had told the teams that its staff would wait to disclose the precise course until just two hours before the start—at 4:30 A.M. the morning of the race. Red Team was getting good at creating routes through the desert. So what if they changed strategies? What if, rather than focusing on creating a map that featured a pre-driven route along every single conceivable trail through the desert, they instead became really good, and blindingly fast, at teaching Sandstorm to drive a single trail?

Rather than a perfect map, they thought, why didn’t they focus on creating a single, perfect route? One they could plan out in the two-hour span between the time DARPA disclosed the approximate course and the start of the race? The old way involved using the maps and the route planners during the months before the race to effectively pre-drive every single road through a desert that covered a territory of fifty thousand square miles. This new way involved focusing on a single 150-mile path that the planning team would examine in fine detail—and doing it in the 120 minutes that passed after DARPA disclosed the race route.

From that moment on, one part of Red Team focused on executing the second epiphany. In the old high bay in the Planetary Robotics building, about a dozen members rehearsed exactly what would happen after DARPA handed over the route in a computer file at 4:30 A.M. The file would feature a series of about 2,500 GPS waypoints, which everyone referred to as “breadcrumbs,” spaced about a hundred yards away from one another, tracing out the course in a dot-to-dot fashion. The dozen members of Red Team’s planning unit would leap into action. One would feed the file into a software program that used the Mojave map’s cost estimates to build a more precise route, with many times more breadcrumbs than DARPA’s route network definition file (RNDF).

But Urmson, Whittaker and their team didn’t trust the route calculated by the planning software. It had been known to send Sandstorm on journeys that went over ridges, into ditches or through wire fences. So a team of editors would divide up the course into sections and then, using computers, virtually go over every yard of the computer-calculated race path to make sure the software hadn’t made any mistakes. Once the human editors were done correcting the course, they’d reassemble it into a single route and upload it to Sandstorm, to execute on the race course.

Still, by January 2004, just two months before race date, Sandstorm had not yet gone fifty miles on its own. One thing causing Whittaker and Urmson anxiety was the disconnect between where they were testing Sandstorm and the race course. They were testing the robot on the frigid shores of Pittsburgh’s Monongahela River. The race would be held in the Mojave Desert. Would the change in environment pose a problem to Sandstorm?

In February, Whittaker arranged for some of the team’s key members, including Urmson, Peterson and Spiker, to accompany Sandstorm to the Mojave Desert to refine the robot’s capabilities. (Sandstorm actually made the trip in a fifty-two-foot enclosed semi-trailer.) The final part of preparations would happen at the Nevada Automotive Test Center, an enormous swathe of desert where companies from all parts of the automotive sector, from tire manufacturers to transmission firms, tested their products in the harshest desert terrain available.

In Nevada, Urmson’s team worked exclusively on Sandstorm. Write code, take Sandstorm out to test the code, watch for mistakes, take note of the mistakes, write code. They repeated the cycle without regard to clocks or arbitrary separations of day and night. Two, three days at a time they worked without sleeping, fueled by Mountain Dew, Red Bull and junk food, and then, when they were too exhausted to manage to keep themselves vertical, they slept. Sometimes in an RV they’d rented, although the trailer didn’t have enough beds for all of them; others slept on the floor of the test center’s mechanics shop on folding lawn chairs, or in the reclined seats of the SUVs they rented to tail Sandstorm.

Working nonstop, through night, through day, the way they did presented some difficulties. One evening, past midnight, Sandstorm ran into a fence post, wrecking the front bumper, which was necessary to support cameras and radar sensors. The test center’s mechanics building was locked up, of course, but in the spirit of asking for forgiveness being easier than requesting advance permission, Spiker and one of the students scaled the fence and broke into the building, where they welded together an entirely new bumper with thick steel pipe. The thing ended up weighing about two hundred pounds—making it more than able to support the sensing equipment the robot required. “You could probably have driven through a building and not hurt that thing,” Spiker recalls.

One thing they didn’t do much of was bathe. The wastewater tank in their rented RV filled up, and by the time they got around to driving it to the nearest town to empty it, the vibrations from the washboard dirt road into town splashed sewage all over the RV’s interior. Cleaning the mess was so traumatizing that the team outlawed use of the RV’s bathroom. While there were bathrooms available in the mechanics shop, no other showers were available, so the guys went without washing for about six weeks. Then, in mid-February, one of their computer sponsors, Intel, invited the Nevada members of Red Team to San Francisco, where the computer chip manufacturer wanted to show off Sandstorm at the Intel Developer Forum.

By that time, Sandstorm had managed a speed of 49 mph and an autonomous run of a hundred miles. The guys were excited about the progress they’d made. But the robot still had its mechanical idiosyncrasies. It was apt to see obstacles that weren’t there, or miss obstacles that were, or even misinterpret pre-programmed commands. What if something like that happened while Sandstorm was onstage at the conference?

The following morning, an audience of hundreds watched the autonomous vehicle creep out onto the stage, apparently thanks to the benefit of high-tech sensors, engineering and computers powered by “Intel inside.” The crowd cheered in response. The applause felt good to the Red Team members present. Here they were at a Silicon Valley event being treated like celebrities. The recognition validated their sacrifices and the worth of the project. It also made the team thankful that no one realized that during the onstage demonstration, a Red Team member had hidden in the space under Sandstorm’s steering wheel, prepared on a moment’s notice to slam his hand on the brake pedal if the massive robot threatened to roll off the stage into the crowd.

On Friday, March 5, 2004—eight days before the race and just three days to go before the qualifying events—Chris Urmson rose early in the morning, put on his usual uniform of a mud-spattered baseball cap, fleece sweater and worn jeans, laced up his running shoes and decided that today would be the day to stage Sandstorm’s culminating test challenge.

Urmson, Peterson, Spiker and the rest of the Nevada squad tested Sandstorm in the worst conditions they could imagine—frequently, along sections of the trail the old Pony Express had followed more than a hundred years earlier. “Red is really gung-ho about testing hard,” explains Peterson. DARPA had said its route would be about 150 miles. The longest run Sandstorm had made was a hundred miles. But with the race a little more than a week away, everyone on the team was hoping for a longer run to boost their confidence.

The goal was just like the race: 150 miles in ten hours. The route amounted to a flat oval, about two miles around. While they prepared Sandstorm, Urmson and Peterson tinkered with a new part of the software: a component of the speed-setting module designed to slow down the robot when it approached a curve. The new code was designed to allow Sandstorm to drive more quickly on straightaways.

The code worked wonderfully. During a few warm-up laps, Sandstorm managed to get up to 49 mph along the straightaways and then the new algorithm slowed it down as the robot headed into the curves. In fact, as Urmson and Peterson watched the robot, they wondered whether it slowed Sandstorm too much. An adjustment to the algorithm during a refueling break seemed to improve things. On the first lap they watched as Sandstorm cruised into a curve, slowed a little bit and then accelerated through the curve’s exit. At the end of the second lap, Sandstorm was heading fast into what Urmson would later describe in his field test report as a “soft S-curve” to the left. The right-side tires drifted off the road into deep sand, and when Sandstorm tried to correct things, to get back on the track’s packed-down dirt, it steered too hard to the left. The right-side tires bit into the soft sand. The left-side tires came up off the road. Behind, in the chase car, Urmson watched, horrified, as Sandstorm tipped up and over, and came to rest upside down—right on top of the e-box and the gimbal housing all the vehicle’s most sensitive equipment.

The robot had been designed to insulate the box’s components from being damaged in all sorts of accidents. Front-end collisions, rear-end collisions—pretty much any collision that happened on the ground plane, Sandstorm would be able to withstand just fine. But the robot had one fatal weakness: a rollover. Because Humvees sat comparatively low and flat, their geometry made rollover accidents almost impossible.

Unless you were testing a robot Humvee in the Mojave Desert, apparently.

A History Channel crew had come out to film the test run. They rushed out onto the track with their cameras and shoved one into Urmson’s face, asking him to list the damage. Urmson looked at the wrecked robot the team had spent the better part of a year engineering: at the crushed gimbal, the compacted GPS antennae, the flattened e-box and the connecting rods bent out of shape. And he let fly with the expletives that made him one of the few people to ever have to have been bleeped by the History Channel. “Shock and disbelief,” Urmson says when asked to describe his reaction, more than ten years later. “But mostly disbelief.”

Disbelief, because they had felt like they’d been making such great progress. Disbelief because they were just days from the qualifiers. Disbelief because this had happened on the second lap of a two-mile track that Sandstorm was supposed to drive for seventy-three more laps.

Most of the crew figured Red Team was over. That they’d never be able to repair the robot in time. Somebody called Pittsburgh to inform everyone else about the accident. Red’s assistant Michele Gittleman took the call. She recalls sobbing when she processed the news.

Maybe a crew led by someone other than Red Whittaker would have given up. But Whittaker didn’t even consider it.

At the Nevada Automotive Test Center, Urmson, Peterson, Spiker and the rest of the team attached the four-by-four chase vehicle to Sandstorm with the help of some nylon webbing and managed to flip over the robot. Spiker, the most mechanically minded among them, went over the engine to look for problems. The engine was flooded with diesel fuel but aside from that, everything looked okay.

The other guys assessed the electronics equipment. The GPS units were toast. The gimbal suffered the worst impact and would need to be completely rebuilt. The main LIDAR unit was irreparable. Luckily, an extra gimbal and LIDAR sat in storage back in Pittsburgh. They towed Sandstorm to the mechanics shop and for three nearly sleepless days and nights they worked to fix everything they could. And they nearly did it.

The race was March 13, 2004. Heading into the qualifiers the week before, at the California speedway in Fontana, the GPS system worked, which meant the robot could locate itself in the world. The sensors were active, which enabled Sandstorm to perceive obstacles. The computers could calculate the trajectories required to follow the path set out by the Red Team mapping crew. The only problem? “We had no time to calibrate,” recalls Whittaker. Which meant Sandstorm viewed the world through a distorted lens.

Think of each sensor as its own individual eyeball. You are able to see one version of reality because your brain is able to amalgamate the view from your two eyes into a single picture of the world. Sandstorm amalgamated the information from four different LIDAR units plus the stereo-camera system. Ensuring that the robot’s sense of the world resembled the actual world required calibrating the individual sensors—a time-intensive process of trial and error. “Think of calibration as alignment,” Whittaker explains. “Even in the car shops they align your headlights, right? And when there are multiple sensors that will fuse data into a common model, it’s important that they’re all aligned. If you just bolt it back together you’re creating a kind of Frankenstein, and maybe it’s a little cross-eyed.”

And so the cross-eyed Frankenstein’s monster limped into race week, belching diesel exhaust, dented and scratched, but otherwise intact. Red Team would compete against twenty other entries from across the United States. In the qualifiers each of the twenty-one robots would have to navigate a mile-long obstacle course to progress to the main event.

Soon after his arrival at the California Speedway, Urmson wandered around to see what he could learn about his competition. He saw Doom Buggy, created by the only high school admitted to the competition, Palos Verdes, near Los Angeles. An undergraduate from UC Berkeley, Anthony Levandowski, led the team behind the only two-wheeled entry, a robot motorcycle that was able to balance itself with the help of a gyroscope. UCLA’s entry, the Golem Group, was led by a guy named Richard Mason, who had seeded his project with $28,000 he’d won on Jeopardy!

On the other end of the spectrum were the professional teams, which were affiliated with various engineering-focused corporations. An inventor named Dave Hall had created an autonomous Toyota Tundra pickup truck that was notable for driving smoothly with a stereoscopic camera setup—it didn’t use any LIDAR at all. From Wisconsin, the makers of Oshkosh Trucks entered a six-wheel-drive, 32,000-pound fluorescent yellow behemoth with the imposing name of TerraMax. Louisiana’s Team CajunBot also used a six-wheeled vehicle. A fraction the size of the Oshkosh entry, it was based on an all-terrain vehicle more commonly used by the state’s hunters to navigate the bayou.

Urmson, Peterson, Whittaker—all of them wandered the event, talking to people just as technically minded as they were. It quickly became apparent that Red Team was among the biggest of the teams. Popular Mechanics gave them seven-to-one odds to win, highest of all their competitors’. The front-runner status positioned Red Team as the entry everyone else wanted to beat. Urmson had posted on the Red Team website a photo of Sandstorm after the rollover. Now, as he wandered the raceway, talking to the leaders of other teams, Urmson spied the photo on numerous computer monitors. Some other teams had made it their wallpaper—as motivation.

Adding to the excitement was the fact that DARPA’s public relations team had arranged for reporters and television producers from across the nation to visit the raceway. Urmson and his teammates had toiled for months in obscurity. The accolades they received at the Intel event had been nice, but the more common reaction to their work was incredulity. “A car that drives itself?” people would scoff. To many, it sounded ridiculous. The presence of the reporters going around interviewing anyone available reminded the competitors that their work was important. Important enough for the U.S. government to put up a million-dollar prize. Important enough, possibly, that it might save the lives of U.S. soldiers fighting in distant theaters of war.

On the morning of March 13, 2004, the start of the race was one of the most exciting moments Chris Urmson had ever experienced. The robots were lined up in their starting chutes. Media and military helicopters hovered in the sky. Grandstands supported hundreds of spectators, each of them getting whipped by the desert sand, and over it all, Tony Tether’s amplified voice marked the momentous event.

“We’re thirty seconds from history,” shouted the DARPA director into the microphone. “All right, ladies and gentlemen, boys and girls, the bot has been ordered to run, the green flag waves, the strobe-light is on, the command from the tower is to move!”

Because Sandstorm had performed best in the qualifiers, it had the honor of starting first. The big Humvee rolled slowly out of its chute. “Ladies and genetlemen, Sandstorm!” Tether cried. “[An] autonomous vehicle traversing the desert with the goal of keeping our young military personnel out of harm’s way.”

The first complication in the race course was a leftward turn. Its inside edge was marked with some scrubby vegetation, and its outside edge, with concrete jersey barriers protecting spectators from the robots. Sandstorm followed the road perfectly throughout the curve and accelerated once it headed out on its straightaway.

While it was still in view, Sandstorm ran over a hay bale. Urmson winced. But the big off-road vehicle just kept on going. Soon, the Red Team couldn’t see their robot at all. No one had thought to provide the teams with a video feed of their vehicle’s progress. All they could do was settle in and wait to hear reports issued back to the start from helicopter-borne observers and other officials set along the course.

Soon, the other entries headed out: A team called SciAutonics II. Then CajunBot rolled its six wheels from the starting chute and drove straight into a jersey barrier. Team ENSCO’s robot, based on a Honda ATV, wandered from the road just past the turn, flipped over on its side and was out of the race just two hundred yards into the event.

Palos Verdes High School’s autonomous SUV also ran into a jersey barrier. And then came the most curious of the entries: Anthony Levandowski’s autonomous motorcycle. Levandowski pulled it up to the starting line, activated the gas-powered motor, stepped away—and watched, brokenhearted, as the motorcycle immediately tipped over. As Levandowski would discover later, he’d forgotten to activate the gyroscope that kept the motorbike balanced. His race was over.

Minutes later, Red Team heard from a race organizer that something was wrong with Sandstorm. The hay bale the robot had run over just after the start turned out to reflect an ongoing problem. Perhaps because its sensors hadn’t been calibrated properly, perhaps because the main LIDAR’s replacement unit scanned at a much slower rate than the original, Sandstorm consistently appeared to think that it was a foot or two to the left or right of where it actually was. The Humvee drove over a fencepost, then another and a third. Some miles later, the vehicle swung itself into a curve, a particularly tricky one given the inside edge was separated from a steep drop-off by only a knee-high berm. As Urmson and Peterson intended, Sandstorm slowed down as the road turned. But the robot was a foot or two to the left from where it should have been. As a result, the left-most tires climbed up the berm, then dropped down the steep inside ledge. Sandstorm was now stuck on its belly—what Urmson called “high-centering.”

Things quickly grew worse. Sensing Sandstorm wasn’t moving, the speed control system kicked in, directing more power to the engine. One of the tires hanging over the other side of the berm was situated just high enough off the ground that it could still touch the Mojave sand. The friction heated up the rubber until it smoked and eventually burst into flames. The robot’s progress was over 7.3 miles from its start.

The media used Sandstorm’s flame-out as a metaphor for the entire event. The number-two entrant, SciAutonics II, also got stuck on a low hill of earth. Dave Hall’s Toyota Tundra became confused by a small rock. The UCLA entry, Golem Group, stalled out when a safety device prevented its engine from accelerating enough to get up an incline. And TerraMax, the 32,000-pound monster truck known for its brute force approach, ended up halted when a pair of tumbleweeds it incorrectly considered immovable obstacles blew ahead and behind it. And those were the best-performing vehicles.

The result put DARPA director Tony Tether in a tough spot. At the other end of the race course, in Primm, Nevada, was a tent full of reporters who had traveled across the country to file stories on the race winner. Tether figured he was going to get killed by the press—an expectation that proved right. “DARPA’s Debacle in the Desert,” went one headline. The gist of the stories portrayed DARPA as an out-of-touch government bureaucracy that had wasted money staging a fool’s errand. So to distract them, Tether took the stage and announced a second race, to be held in a year or so, with a doubling of the 2004 race’s purse, to $2 million.