Thus came the idea of the DARPA Urban Challenge, which would be staged in a city landscape, rather than a desert. Tether announced the event in April 2006, setting the date for November 3, 2007, and soon saw a globally diverse field of eighty-nine teams registered to compete, less than half the number of the previous contest, perhaps because this version of the challenge was perceived to be much more difficult.
Some of the format changes seemed designed to stifle the approach that Red Whittaker’s team had pioneered in the first and second races—the one that had used a group of human map techs to essentially pre-drive the race route for the robots. DARPA planned to sprinkle the course with moving obstacles, namely, other automobiles driven by Hollywood stuntmen and professional drivers. As well, many different teams would be navigating the urban environment simultaneously.
The key to this event was satisfying the objectives DARPA wanted the robots to execute throughout the course—and DARPA wouldn’t disclose that information until five minutes before start time. In fact, DARPA was so secretive in the lead-up that for a time they even declined to disclose which state would host the competition. “We knew it was going to be cold outside, so they’d probably do it somewhere that had warm weather,” recalls one participant. “But that was the only thing we knew … They didn’t want anyone pre-programming the system. They wanted some element of intelligence and route planning and control in the robot.”
The rules required the robots to drive sixty miles in six hours through an urban environment while obeying the rules of the California Driver Handbook. The challenge would require that the robot navigate a standard North American parking lot well enough to be able to maneuver into an available space. While neither pedestrians nor cyclists would be allowed on the course, the robots would have to navigate one of the most difficult elements of driving for human operators—deciding how to proceed through an all-way-stop intersection where other drivers have arrived at about the same time.
“The Urban Challenge was much harder, in terms of what the vehicle needed to do,” recalls Urmson. “The algorithmic steps we’d taken for the first two challenges were predicated on the world not moving. Once other things start moving, it’s nowhere near as easy.”
The inspiration here was to automate the operation of battlefield convoys. A military truck in Afghanistan or Iraq is transporting food to a distant village. An IED explodes somewhere ahead, and the automated convoy would have to navigate around the disturbance without running into any medics, civilians or other members of the convoy. That’s about as dynamic an environment as is possible.
There was no question that Carnegie Mellon would enter the race. There was some question whether Red would lead it. Previously, Whittaker’s Red Team had been a take-all-comers effort populated by undergraduates, volunteers, grad students or the odd full-time employee of Whittaker’s Field Robotics Center. But this time around, DARPA was pledging a million dollars of research funding to a selection of the best-run teams. Carnegie Mellon was one of the recipients. As well, the stakes felt higher this time. There was the $2 million prize. In addition, Carnegie Mellon was competing for its reputation as the nation’s top robotics center. It needed to win. “That’s a lot of money and so [the university administration] wanted to make sure we could win it,” Urmson recalls. Ultimately, Whittaker retained his leadership of the Urban Challenge team, but this incarnation featured other senior members of the Robotics Institute faculty—the university’s equivalent of a supergroup.
Some in the university felt that this new group was so different it deserved a different name. Red Team had been appropriate in the past because it had been Red Whittaker’s baby. But this Urban Challenge was the best squad that Carnegie Mellon could assemble. Some of the veteran members of Red Team resented the rebranding. “It made no sense to us,” recalls Michele Gittleman, Red’s assistant at the time. “Everyone knew Red Team. The brand had already been established. We had hats, T-shirts, jackets.” Nevertheless, to reflect the sense that this was a new effort, fully backed by the university, the Carnegie Mellon team rebranded itself as Tartan Racing, in a nod to the nickname given the university’s sports teams, itself a reference to founder Andrew Carnegie’s Scottish heritage.
This latest effort was going to require a lot more than a million dollars to fund. So in 2006, Whittaker came and visited me at the General Motors Technical Center. “Why do you think you’re going to win this challenge?” I asked him. “Dust,” Whittaker replied, explaining that the robots created by many other teams treated dust clouds as impermeable obstacles, unable to be driven through, while Carnegie Mellon’s software and sensors were able to correctly tell that dust represented no obstacle at all. Although dust represented a comparatively small factor at the actual Urban Challenge, the answer convinced me. I liked Whittaker from the moment I met him. With his military bearing and his eminently American positivity, his confidence that ingenuity and hard work could solve any problem if you persevered, he struck me as a throwback to an earlier era of technical innovators, the sort of people who pioneered the automobile a hundred years before. I arranged for GM to back Tartan Racing in numerous ways. The corporation would end up providing Whittaker’s team with $2 million in support, making us the team’s lead sponsor. We also provided the services of some of our top engineers, and embedded one of them, Wende Zhang, with the Tartan Racing team in Pittsburgh. Tartan Racing’s vehicle would become a 2007 Chevy Tahoe dubbed “Boss,” after my long-ago predecessor as General Motors’ vice president of research and development, Charles “Boss” Kettering. Other sponsors also donating funding included the construction-equipment maker Caterpillar; the auto-parts supplier Continental; and Applanix, a manufacturer of GPS systems.
One of the new personalities on the Tartan Racing team was Bryan Salesky, who ran the software team. Salesky had been working at a Carnegie Mellon Robotics Institute spin-off called the National Robotics Engineering Center. Whittaker cofounded it in 1994 with $2.5 million in funding from NASA, to commercialize the technology created at CMU’s robotics department. The place is located in a nineteenth-century former iron foundry in the Lawrenceville area of Pittsburgh, on the shores of the Alleghany River. Its job was to partner with such companies as John Deere and Caterpillar to develop commercial projects like a self-driving harvester or an autonomous excavator. Shortly before work began on the Urban Challenge, Salesky was working on an autonomous navigation system for the U.S. Army. But the project wasn’t going anywhere, and Salesky was becoming frustrated with the slow rate of progress that he thought endemic to working directly with the government.
Just twenty-six years old, Salesky was, on the surface, a bit of a strange fit for such a senior position in Whittaker’s crew. Guys like Spencer Spiker, Kevin Peterson and Chris Urmson are equally at ease with welding torches and air impact wrenches as LIDAR sensors. Salesky is more at home in a computer lab than a mechanic’s shop. But he hit it off with Whittaker’s guys, particularly Urmson, likely because the two men shared a Midwestern distaste for pretense.
In 2006, as teams across the country geared up for the Urban Challenge, other figures who would become main characters in the mobility disruption began bumping up against one another. For example, Dave Hall and Anthony Levandowski. Hall, then fifty-five, was an inveterate tinkerer and self-trained inventor. He first gained notoriety in the world of high-end audio, having famously built his first amplifier at the age of four. Living then in Connecticut in a family with lots of technical heritage—Hall’s father designed atomic power plants, and his grandfather was a physicist—he already knew how to read electronics schematics when he went to college for mechanical engineering. While there he invented a version of a tachometer, a device that measured the rotational speed of things like wheels and propellers. Licensing the patent from that invention gave him enough income that, out of college, rather than getting a job, Hall moved to Boston to set up his own little technical shop. For a time, he lived by monitoring the world of government research contracts and building prototypes for the big defense firms. In the seventies, Hall invented a subwoofer, a type of speaker that provided clearer bass tones in stereo systems. With $250,000 lent to him by his grandfather, Hall moved to California and set up a company with his brother-in-law to manufacture the subwoofer.
That was in 1979. By the turn of the millennium Hall’s company had sixty employees and a few million dollars in annual sales. Hall made a good living, but he was bored. Building remote-controlled robots that competed to “kill” other machines on BattleBots, the television show, occupied him for a time. (One of his constructions, Drillzilla, relied on the blade from a circular saw as a weapon.) He competed in the first DARPA Grand Challenge with a Toyota Tundra entry notable for its use of a stereoscopic camera setup, rather than LIDAR, to sense the road. In the interim between the first and second races, Hall became fascinated with LIDAR’s potential. Tinkering as he was wont to do, Hall determined a way to cram sixty-four lasers on a single device, more than anyone ever had before. But the big innovation was that Hall’s LIDAR spun. Previous laser range finders had stayed static, shooting lasers to get a limited field of view, a little like the way a human sees the world only in front of his or her eyes. By mounting the device atop the vehicle and engineering the LIDAR so that it revolved ten times a second and 360 degrees around, Hall’s mechanism provided a complete portrait of the area around the vehicle. The new technology in effect created a three-dimensional scan of the world. It helped Hall’s robot progress along the second Grand Challenge route faster than most of its competitors, although a mechanical failure prevented it from finishing the race.
Once DARPA announced the Urban Challenge, Hall saw that his LIDAR would be even more valuable, because the 360-degree field of view that his device provided would help the robot detect oncoming vehicles in all directions. So he set up a manufacturing operation within his subwoofer company, Velodyne, and hired as a salesman one of the brightest minds from the first two races. The person he hired was Anthony Levandowski, the thin, six-foot-seven-inch UC Berkeley grad student behind the first challenge’s Ghost Rider motorcycle.
Which is how Levandowski came to be in Pittsburgh at the old Coke Works test facility—now referred to by most as Robot City—one day in late 2006, about a year before the race. Tartan Racing had bought one of Velodyne’s LIDARs—a significant investment at about $75,000 a pop—and Levandowski flew out to Pennsylvania to help Urmson and team install it. He set it alongside the other sensors that sat on the metal latticework on the roof of Boss, the Chevy Tahoe. Levandowski anchored down the sensor, activated it, and the numerous computer scientists and engineers who had gathered to witness this moment watched as the device generated the rotational momentum it required to work properly. On a nearby computer screen appeared the ghostly dot matrix that formed the device’s output. It was impressive—a million data points that could recognize up to 120 yards away everything from parking curbs to people’s faces.
Then something came loose on the LIDAR, and the rotational momentum flung a counterweight across the room—hard. Only luck prevented the wayward schrapnel from injuring someone. There was a shocked silence, and then came Levandowski’s voice.
“We’ll fix that,” said the gangly engineer.
Levandowski would become well known years later as the central figure in an important lawsuit between Waymo and Uber. Brilliant and ambitious, Levandowski demonstrated a tendency to get himself into situations others would characterize as conflicts of interest. He was often in demand at the smartest, highest-performing groups working on the most intriguing projects. Velodyne would ultimately sell its LIDAR to at least seven of the Urban Challenge competitors, including two each to the Carnegie Mellon and Stanford teams. Around the same time Levandowski was selling this integral piece of technology to as many teams as he could, however, he also was advising the Stanford team for the Urban Challenge. There’s no indication he did anything unethical at this point; Stanford apparently knew about his role with Velodyne, but it is easy to see how this activity could be perceived as a conflict of interest. And just to make all this even more incestuous, Google’s Street View played a key role in how it all went down.
By 2006, Thrun was itching to do a start-up. The atmosphere of Silicon Valley played a role, as did the Stanford AI prof’s developing relationship with Google cofounders Larry Page and Sergey Brin. But what should the start-up do?
Thrun was fascinated with the data that he and Montemerlo had collected while testing Stanley for the second Grand Challenge. Teaching Stanley to drive, motoring in the Touareg around the Mojave Desert, Thrun and his teammates had struck on the idea of fixing cameras, pointed in several different directions, on the roof of the vehicle. Not as sensors—rather, the imagery helped them re-create the circumstances that triggered bugs in Stanley’s programming. Over time, Thrun realized how interesting it was to just go through the imagery collected by the multidirectional rooftop cameras.
That year, Thrun happened to be teaching a class at Stanford about computer vision. He assigned the class’s most brilliant student, Joakim Arfvidsson, the task of creating a program that could easily stitch together the camera footage in such a manner that it provided the illusion of an unlimited field of vision, which, Thrun figured, would provide the impression of actually being at the original location. “Joakim recorded a street in San Francisco,” says Thrun. The resulting computer program gave the feeling of actually standing on the street. “You could look up, look down—it was completely amazing.”
Thrun gave Arfvidsson an A-plus in the class. During the summer of 2006, alongside his oversight of Stanford’s Urban Challenge team, Thrun assigned a second team the task of building a version of the street-visualization software that could work on a cellular phone. The team included Hendrik Dahlkamp, Andrew Lookingbill and Arfvidsson. Thrun installed his good friend Astro Teller, whom he knew because Teller did his PhD in artificial intelligence at Carnegie Mellon, as the start-up’s CEO.
In the first months of 2007, Levandowski also joined the team. Preparing to show the technology to venture capital firms for seed funding, Thrun aimed to stage a really impressive demo—he wanted to be able to place the VCs on any street in San Francisco, which required driving their camera rig along every street in the city. Levandowski was the one who determined how to do that quickly. He figured out that rental cars were really cheap if you hired them by the month. Next, he procured large numbers of drivers by advertising on craigslist. It took about two weeks to complete the map. “He was a really great person, I would say, to get shit done,” Thrun recalls.
March 2007 was the month Thrun made his approach to VCs for funding. The effort became all-consuming. “Just strategizing with these venture capitalists becomes this amazing, intense game,” recalls Thrun. “My life is completely taken over. I have no social life—my wife thinks I’m a moron.” The effort paid off, though. Two of the valley’s top VCs, Sequoia Capital and Benchmark, were both interested. Thrun set the bidding for a Sunday, April 8, 2007. Soon the bids were climbing: a $5 million round of seed funding, which became $10 million. Then $15 million.
That evening, as he mulled over which venture capital firm to choose, Thrun invited himself over for dinner to Larry Page’s place. Sergey Brin showed up. The men discussed Thrun’s technology, which he called VueTool. Page already had sponsored something similar. Some time before, he, Brin and Marissa Mayer had gone out and taken some footage that they then stitched together. (In fact, according to Thrun, the idea of immersively stitching together camera imagery so that it was possible to click through it had been invented in 1979 by an MIT scientist named Andrew Lippman.) And Page and Brin had a similar project happening at Google, one run by a guy named Chris Uhlik. After dinner, Thrun took Brin and Page to his office at Stanford to demo the footage of the San Francisco streets. The Google cofounders were impressed; they saw that Thrun’s team had accomplished a lot more, much more cheaply and in a lot less time, than their people had. For example, Google’s in-house Street View team was using custom-built camera rigs that cost $250,000 each, according to Mark Harris’s reporting in Wired. Thrun and Levandowski, Harris writes, were getting images of similar quality using a setup of off-the-shelf panoramic webcams that cost $15,000.
The next day, Google’s head of mergers and acquisitions called Thrun, who agreed to sell the VueTool technology to Google. As part of the deal, Thrun, Levandowski and the rest of the team joined the company to accelerate the Street View project. “We got fairly lavish bonuses,” Thrun explains. In following with the Red Whittaker approach of setting an ambitious goal to motivate his team, Thrun set up an arrangement with Google that would trigger another bonus payout if Thrun, Levandowski and the team were able to map a million unique road miles for Street View. Using the method that Levandowski pioneered, with lots of cars, Thrun and his team ended up meeting the milestone in just seven months.
Thrun’s obsession with meeting the Street View goal meant the day-to-day work on Stanford’s entry to the DARPA Urban Challenge was led by Mike Montemerlo. On the other side of the country, Montemerlo’s former officemate Chris Urmson led the day-to-day work on Carnegie Mellon’s robot vehicle. Urmson approached his effort on this race like it was the most important quest of his life. Every so often he recalled the conversation with his wife, Jennifer, nearly four years before, when he promised that he’d just do the first desert challenge, before going off and getting a real job outside academia, where he could start making a good living to support his growing family. (He and Jennifer had since had a second boy.) Sandstorm’s rollover before the first race had provided him with the sense that he could have won it, if not for that accident. H1ghlander’s mystifying mechanical issues in the second DARPA challenge provided a similar sense of the team’s tantalizing proximity to victory. This urban event, Urmson knew, likely represented his last chance at victory. His last, best shot.
Leading up to the race, Team Tartan often discussed the rules. How difficult would DARPA make this challenge? Did DARPA even want a winner? It would be a simple enough matter to make the race so tough that no team could ever win. That would be the most cost-efficient option. If the government’s intention was to outsource development of autonomous vehicles, to prove the possibility of self-driving cars while investing the minimum amount of money, then one way to do that would be to stage a race that prompted universities and research centers all over the country to work on the problem, while making it so tough that DARPA wouldn’t have to actually hand out the prize money.
By this point, building an autonomous robot had become nearly routine. Transforming a Chevy Tahoe into the self-driving Boss resembled the maturation of a human being, in some respects. The vehicle started blind and dumb, unable to sense, to navigate, to move on its own. Then Urmson and his team installed sensors—the LIDAR, the radar—as well as the computer processors. In their earliest tests the team taught the robot not to walk but to drive, supplying it with a list of GPS waypoints, similar to the sort that earlier generations of CMU vehicles had been required to follow in the desert challenges. Once the robot could trace the dots of a mile lap around the grounds of the old steel mill, the team set up longer waypoint-finding tests. In November 2006, a full year before the competition, Boss completed a fifty-mile route, achieving speeds of 28 mph.
In parallel to Boss’s mechanical testing, Salesky’s programming team worked to incorporate perception and planning systems into Boss. The robot could understand the input it was getting from its rudimentary eyes, in the form of its Velodyne LIDAR and radar sensors. In December, Boss achieved a multi-checkpoint mission, running at night along the cold Pittsburgh riverside. Tartan Racing also coded rules that would dictate how the vehicle would behave during the situations it encountered while driving. Intersection handling was an early module in which the programmers created a set of directions for the various situations the robot might encounter at an all-way stop. What if Boss arrived first at an intersection, followed by someone else to the right? What if Boss arrived second? Salesky’s team created rules for each scenario.
Around the same time, Tartan Racing integrated its behavioral software with the hardware. A synchronization board regulated the timing of the information coming in from each disparate sensor, allowing Boss’s computing cluster to build a simulation of 3-D reality, the same way human drivers use their eyes and ears to build a model of the world in their heads. Another step involved predicting the behavior of other objects. For cars to be truly autonomous, they would have to be able to anticipate the behavior of practitioners of many different modes of urban transportation, from pedestrians to cyclists, skateboarders and scooter riders, among many others. But DARPA had told its teams that this race would happen in a vastly simplified environment; the only moving objects on the suburban course would be other automobiles. That made things a lot easier for the men and women coding the software, because it meant that Boss had to understand only a single set of behaviors—a tendency to move only forward and back along curvilinear lines, for example. Everything else, to Boss, was a stationary object.
As 2006 gave way to 2007, I flew to Pittsburgh to visit Red, Urmson, Salesky and the rest of the Tartan Racing team members at Robot City, including the GM engineers working with the team. I can remember touring the team’s workspace in the winter, amid these post-industrial ruins, the railroad roundhouse, the trailer set on the frozen ground, and being struck by how chilly the offices were. Everybody walked around in winter coats with knit caps on their heads. You could see your breath. These guys are really living lean, I thought. Rather than spending their budget on creature comforts, it impressed me that they were reserving it for research and technology.
Despite the Spartan feel, Tartan Racing’s headquarters seemed glamorous to me. I was a top executive in a major corporation, managing a budget in the billions of dollars at GM, but a part of me envied these young men. They weren’t the sort of people who crunched numbers in an ivory tower. They were “salt of the earth, roll in the mud, get your fingers dirty” engineers—who just happened to believe they could change the world. Shielded from the university bureaucracy by Whittaker, unencumbered by the sort of corporate red tape required by General Motors, these guys were getting things done that GM never could.
I watched as Whittaker, Urmson and the Tartan Racing team tested Boss in the sort of situations that had been modeled using computer simulations only weeks before. Boss’s first challenge at the demo I witnessed was a three-way intersection. The rules the programmers had coded into Boss’s digital brain guided the robot to yield to other vehicles that had arrived before it, and when it was Boss’s turn, the robot rolled through without a problem—an enormous relief to the team. As an additional feature of complexity, the team incorporated bad drivers into the equation. In one such test conducted at Robot City, Boss arrived at an intersection after a white American-model sedan, followed by a third vehicle—the second Humvee donated by AM General before the second Grand Challenge. The white sedan went first. Boss inched forward—and then the Humvee sped through the intersection, completely out of turn. Rather than rolling into the Humvee, Boss paused—exactly the appropriate action.
I was so excited I asked to go for a ride in Boss, which earned me some looks. The team didn’t seem certain it was a good idea. But I insisted, and moments later, I was squeezing myself into the only bit of interior cockpit space not occupied by computer equipment or batteries. Soon I grasped how different this robot was from the sort of vehicles my team designed at GM. Boss accelerated toward a stop sign and braked at the last minute. It careened around a turn and spun its tires over potholes and rocks without slowing. Every motion happened with a jerk or a slam. Just a minute or two into the ride I felt carsick for the first time in years. Now I understood their hesitation. Boss wasn’t designed for human occupancy, Whittaker explained soon after my ride. Rather, it had been designed for the express purpose of winning DARPA’s third race. The robot had been programmed to accelerate aggressively, and brake hard when the situation warranted. The herky-jerky behavior made humans motion sick—but it also made Boss fast.
After an open house inviting the public to witness a demonstration of Boss’s autonomous driving, the Tartan team loaded Boss into a tractor trailer and drove it to Arizona, where my staff had arranged for the robot to be tested at the GM Proving Grounds at Mesa. The warmer weather and the wide-open space of GM’s test tracks allowed Boss to begin learning how to behave in parking lots—which DARPA had said would be a key part of the Urban Challenge. The team modeled left-hand turns into traffic, a difficult proposition for many human drivers. And when the temperatures warmed up in the spring, Boss returned to Pittsburgh to prepare for a visit from DARPA, during which the Urban Challenge’s program manager, Norm Whitaker (no relation to Red), would monitor the sentient Chevy Tahoe as the robot conducted a series of tests. How the robot performed would dictate whether Tartan Racing would be one of the teams that remained in the race when DARPA staged the next round of eliminations. The resulting shortlist would compete at the so-called semifinals, the national qualifying event at the end of October 2007.
During the site visit, Boss had to pass four different tests. These challenges were conducted on Robot City’s quarter-mile track. Norm Whitaker and a crowd of about a hundred, including media and sponsors, watched as Urmson and his team demonstrated that Boss’s emergency stop button could halt the vehicle at a moment’s notice. The robot had to pass through an intersection also used by other vehicles without a collision, which it did. Another challenge involved driving down a street and avoiding a parked car—no problem. “Boss behaved like a good beginning driver,” praised DARPA’s Norm Whitaker. “A real good job.”
At the beginning of August, DARPA director Tony Tether announced the names of the thirty-five finalists invited to compete at the national qualifying event in late October. Tartan Racing was among them, which Whittaker, Urmson and Salesky were expecting. What they didn’t expect was Tether telling a reporter that Boss was not considered to be among the top-five robots.
Consequently, the final months before the qualifiers saw Urmson and Saleskey conducting exhaustive testing of Boss to ferret out any bugs in the algorithms the programmers had constructed—to make Boss as safe and capable a driver as any human being. Some of this testing involved hiding an inflatable model of a car next to the road. Just as Boss was about to pass, a team member would shove the bubble car onto the pavement. Everyone watched to see whether Boss would react in time.
When the robot became expert at dealing with that situation, Urmson escalated the challenge, to using an actual car. One day in late summer the guys were testing in Arizona. Salesky was driving a rental car in front of Boss. “Okay,” Urmson, who was riding shotgun inside the robot, said to Bryan over a walkie-talkie. “I want to make sure the vehicle’s speed control is working. Slam the brakes.”
The first time Salesky slammed the brakes, Boss slowed to a stop. But it wasn’t abrupt enough for Urmson. “Slam them harder,” Urmson told his friend. Moments later Salesky swerved in front of the robot then jammed his foot down on the brake pedal. “Harder,” Urmson beseeched Salesky. “Make the tires skid.”
This was Red Whittaker’s modus operandi. It’s only in the edge cases, Whittaker liked to say, the test-to-failure cases, that you learn about the robot’s capabilities.
Salesky shrugged—and slammed on the brakes.
And Boss ran straight into the back of the rental car. Salesky climbed out and went around to look at the rear bumper. The back of the rental had crumpled like a spent piece of paper. The damage to Boss was worse, because many of its most sensitive instruments, including a pair of medium-range radar units, were bumper-mounted. “The collision blew up about ten grand worth of sensors,” Salesky recalls. Urmson and Salesky stood there, arms akimbo, regarding the damage and shaking their heads. “Why did we just do this?” Salesky asked.
“My fault,” Urmson said.
“No, I should have known not to do it that hard,” Salesky said. He laughs about it, a decade later. “We both knew it was completely unrealistic. The following distance was way too close. There’s moments when you’re testing for so many hours in the field and you’re not really thinking clearly—and that was one of those moments.”
Sometimes, during such testing, a representative from DARPA dropped in on Tartan Racing to make sure that the team continued to make progress, to ensure that its money was being put to good use. “Hey,” the DARPA rep asked Urmson and Salesky during one of these visits. “What are you guys doing to make sure you don’t roll over your vehicles?”
Urmson froze at the question. It was kind of a ridiculous inquiry. The DARPA rep was just needling Tartan Racing’s technical director. The Urban Challenge didn’t feature any off-road territory, so Urmson wasn’t testing Boss in the sort of environments that might lead to rollover accidents. There were no sharp turns in soft sand, for example, which had been the problem with Sandstorm. Nor did Boss navigate arduous trails that featured the sort of ramps that flipped H1ghlander onto its side. “We will not roll over the vehicle,” Urmson said confidently.
And they didn’t. In fact, the worst accident was the rear-ending of Salesky’s rental car. Urmson and Salesky spent weeks racking their brains to think up situations that might mess up the vehicle’s computer algorithms, situations unlikely to happen in real life, let alone in the Urban Challenge. Times, for instance, that all four entrances to an all-way stop would feature the simultaneous arrival of vehicles. How would Boss handle that? Or they’d program in a route and then create an unexpected obstacle along the way, maybe staging an accident that completely blocked the road. Boss would have to calculate an alternate route. There were some near-misses, some last-minute swerves, but for the most part, Boss performed flawlessly. “That was a magical time,” recalled Salesky.
Remember, this was 2007. No one had ever created a robot that could drive in traffic at the speeds that were commonplace on public roads. And as each day passed, Tartan Racing was realizing that it had done it. “The fact that it just generally turned on and worked … it still surprises me,” recalls Salesky. “It was pretty amazing.”