from WIRED: MICHAEL HODGES GEAR | 04.03.15 | 10:00 AM
IF TODD REICHERT cannot regain control of his ultrafast bike, he will be cast at 75 mph into the unforgiving rock and scrub that lies beside the highway in the Nevada desert.
Reichert is familiar with extreme situations: the 32-year-old aerospace engineer and athlete was co-designer and pilot of the first human-powered ornithopter—a craft that flies by flapping its wings—to soar continuously, and the first ever human-powered helicopter to become airborne. These remarkable feats have earned him hundreds of thousands of dollars in prize money and the backing of Google. But today’s attempt to break the world speed record for a human-powered vehicle is going seriously awry.
Reichert wouldn’t be the first man to perish in this dust-blown corner of northwest Nevada’s rough sierras. At an elevation of more than 4,500 feet, the valleys burn beneath treeless peaks. In the 19th century, California Trail pioneers succumbed to thirst and disease here, or were shot by Native Americans. One such bloody skirmish gave a name to the range and the small town that sits beneath it on Interstate 80: Battle Mountain.
Until now, though, no one has contrived to perish in a fibreglass egg. It’s Saturday, September 13, the last day of the World Human Powered Speed Challenge (WHPSC), and the egg in question is the ultra-lightweight speed bike Eta. Before he got in, Reichert, the joint leader of Canadian team AeroVelo, described Eta as “very, very fast—it could be the fastest man-powered vehicle on Earth.” The claim is based on experience: AeroVelo first competed at Battle Mountain in 2011 where the team set a new college record of 116.9 kph (72.63 mph). The following year they entered Bluenose, a bike that reached 125 kph (77.67 mph).A few seconds ago, this ingenious machine was hurtling along a dead straight section of Highway 305, but now it is swerving violently from right to left across the Tarmac. Named after the Greek letter symbolising the efficiency of power supply, Eta uses only four gears to transfer the energy produced by a human pedaling two 26-inch wheels fitted with handmade tubular tires. The vehicle’s ovular appearance results from its two-part fairing—a structure fitted in order to increase streamlining—that has been designed using computational fluid dynamics to have 100 times less drag than modern cars.
The other 11 entrants in the challenge include near-professionals such as the Netherlands’ Human Power Team, a joint effort by Amsterdam and Delft Universities, and the amateur but manically enthusiastic Tetiva team from Russia. The young Dutch team hold the current record of 133.8 kph (83.13 mph) set at Battle Mountain in 2013 by Sebastian Bowier in Delft VeloX 3.
“Where else,” Reichert says on the eve of the team’s 1,677-mile drive from Ontario to Nevada, “will you see 19-year-olds making history?”
But AeroVelo has a wider brief than world records. Established in 2010 by Toronto University graduates Reichert, 32, and Cameron Robertson, 27, the team—from the engineering and aerospace department—wants to demonstrate the potential of human-powered vehicles in a world facing grave energy, transport and climate crises. Over the past few years, Robertson, who is the calmer and more professorial of the two, and Reichert, a fizzing cocktail of dash and intellectual curiosity, have established a production cycle: in winter, with the help of aerospace department PhD Victor Ragusila, they work on the design and planning of a project; in summer they take on a group of engineering and aerospace students and set them the task of building the project from scratch.
Many of the 2014 team—Trefor Evans, Sherry Shi, Calvin Moes, Marc Jutras, Thomas Ulph, Peter Wen and Alex Selwa—have worked on previous AeroVelo projects. Although Reichert will do most of the runs, Evans and Moes too will pilotEta; Reichert’s energies must be rationed if he is going to surpass 133.8 kph. Privately the team are aiming for 140 kph (86.99 mph) and think that it may be possible to reach 145 kph (90 mph). Reaching that speed would smash a record usually broken in tiny increments—in 2013 Bowier broke the previous one by less than a second.
Reichert and Robertson met playing rugby for Toronto University’s engineering department and first worked together in 2006, when Reichert and the university’s Human-Powered Vehicle Design Team attempted to build the first successful ornithopter. Four years of development and several crashes brought them to their final prototype, Snowbird. Made from carbon fibre, balsa wood and foam, it was essentially one very long wing (105 feet—a Boeing 737 has a 112-foot wingspan) attached to a pod in which Reichert operates a customized rowing-machine mechanism. On August 2, 2010, at Toronto’s Great Lakes Gliding Club, Snowbird flew for 9.3 seconds, at a height of 9 feet for a distance of 475 feet—a feat recognised by the Fédération Aéronautique Internationale as the first man-powered ornithopter flight.
It was an achievement that showcased Reichert and Robertson’s ambitions and sealed their professional relationship. “In an engineering and design sense we kind of finish each other’s sentences,” Robertson says at Battle Mountain civic center, the sports-hall-cum-cafeteria where the 12 teams are based for the duration of the challenge. Reichert agrees: “By the end it was quite clear that he was my counterpart, the guy that makes things happen. It was a huge four-year lesson in how to solve engineering problems. It helped us to make innovative decisions when we went on to the helicopter.”This helicopter is Atlas, a four-rotor craft powered by Reichert, this time on a bicycle. The vehicle was responsible for AeroVelo winning a seemingly unachievable science and engineering award—the AHS Sikorsky prize—in 2013. Established in 1980, it required entrants to build a human-powered helicopter that could stay airborne for a minute and reach an altitude of three meters while not straying from the confines of a 10-square-meter box. The $250,000 prize had gone uncollected for 33 years. Atlas weighed 95 pounds: the rotors were made of foam, balsa wood and Mylar. The X-shaped machine had a span of 160.1 feet along the length of each axis and was so large that practice sessions and the attempt on the prize had to be made between matches at an indoor football pitch outside Toronto.
In the run-up to their last attempt, Reichert crashed Atlas on to the pitch’s AstroTurf twice—both times just as he reached three meters. The carbon-fibre rods in the rotor arms were buckling under stress. So, with time running out, the team gambled on shortening the lengths of the arms and risking the overlapping of rotor blades. The rods held, the blades didn’t clash, and, with impatient footballers looking on, AeroVelo won the prize on its 75th flight.
“So many people had said it was impossible,” Reichert says. “Someone did the math to show that it can’t be done. It’s funny how easy it is to show that something’s impossible, compared to how hard it is to make something that was thought impossible.”
Reichert sat upright in Atlas, suspended in a skein of lines, open to the air. In Eta he is recumbent on his back in the carbon-fiber cockpit and completely encased in the fairing. He sees the road ahead on a video screen mounted in the cockpit, which also displays a read-out of his power and speed levels and projected levels. He steers with handlebars that offer only 1.5 inches of grip and is so tightly encased in the fairing that after encountering bumps on the road he emerges with knees smeared with blood. The brakes are directly behind his head and, when applied, throw his full weight on to the front wheel.
The 4.97-mile run begins at a roadside loading yard where thousands of tons of gravel await transportation, and culminates in a 656-foot speed trap where the vehicles trigger a series of cameras allowing a mean speed to be calculated. The bikes then decelerate as they cross a bridge—lined with heavy padding—over a dry creek. They enter a “catching zone” where slowing vehicles are grabbed before they tumble over. The speed trap is marked by orange flags and manned by judges sitting at a table erected in the scrub. The strangeness of the scene is augmented by the presence of three large crows that, throughout the week, habitually sit alongside the judges.
Highway 305 is the main route into Battle Mountain from the town of Austin to the south. For the competition, the authorities have agreed to close it for 20 minutes twice a day—in the early morning and just before sunset. Within these time slots the teams must ensure the wind drops below 3.7 mph if they want their time to be legal, which can lead to a frenetic atmosphere at the starting point. Among the Dutch team, a man on roller skates holds the rear of the bike while another man on skates approaches him at speed. They make contact and, with the first man holding the bike, they set off down the course and release the vehicle. The Russians prefer a frantic ruck around their bike from which it’s catapulted. The team runs behind its creation, shouting and driving pell-mell through the other competitors, somehow leaving them unscathed. AeroVelo’s approach is calm and simple: Evans pushes, Reichert pedals.
On the morning of AeroVelo’s record attempt, Robertson is sitting in the team’s chase vehicle near the end of the course when Eta veers wildly, apparently suffering an uncontrollable loss of steering. If the crows are hoping for roadkill, this could their chance.
Thankfully, Reichert has the physical strength required to maintain Eta’s equilibrium: his thigh muscles are immense outcrops, his arms powerful pistons. This is a man who took up speed skating on a whim after seeing it at the 2010 Winter Olympics and rose to national competition level within a year. His warm-up demonstrates his physical aptitude and mental focus: before each Eta run he completes a series of dance moves, sprints, exercise-bike workouts and target visualisation.
Summoning all his strength, Reichert manages to control the machine and bring it into the catching area. “It was going like a rocket up to 100 [kph],” he gasps as he’s released by the waiting Shi and Ulph. “My cadence was just right and then something went wrong.”
Reichert spends less than a minute sucking air into his lungs before he’s back on his feet. It’s these physical reserves as much as his and Robertson’s intellectual muscle that are key to AeroVelo’s success.
“I understand the human body as an engineering system,” Reichert says as we drive back to the starting point. “So much of this is about how the human fits in a machine comfortably. You need to be able to produce a lot of power, it needs to be stable and yet you have to feel what you do. Of course, if I’d done anything wrong the whole thing would have broken.”
Breakages are always a risk: the week in Battle Mountain had begun badly when Moes slid the machine off the road at 62 mph. Fortunately he was unharmed and the scratched fairing was no more than a filling and sanding job. On Wednesday morning Reichert is working up to full power, posting 61.32 mph on his evening run. As he applies the brakes he hears an ominous “twang”: the sound of spokes shattering. Other teams use tri-spokes or disc wheels but AeroVelo, always looking to cut the bike’s weight, gambled on faired spokes. It transpires that each of the front wheel spoke holes is the wrong gauge and cannot take the pressure. They must be re-drilled.On Thursday—two days before the record attempt—Reichert’s best time is 62.94 mph. Only six sessions remain to reach 86.99 mph, but there is no tension between the pair. “We’ve never had any blow-ups,” Robertson says. “We keep cool. Todd has a lot of physiological issues and I certainly get tired but our minds are always racing—thinking or talking about the next thing.” On Friday morning more spokes explode and Eta leaves the road. It’s a controlled exit, but it necessitates Robertson pulling the van to a halt and the team rushing out to catch Eta before it tumbles into a bush. When he clambers out Reichert has one word: “Spokes!”
AeroVelo’s helicopter—a four-month summer project that ended up taking 18 months—serves as AeroVelo’s difficulty benchmark. “This is nearly as bad as the helicopter,” is heard throughout the week. But the spoke incident provokes something new from a team member: “This is harder than the helicopter.” Robertson looks up and says with a finality that settles the issue: “This is not harder than the helicopter.” But the fear that broken spokes could be the undoing of this attempt has entered the room.
The team sets about rebuilding the wheel in one day and the Friday evening run is fast. Reichert makes 78.47 mph and he’s smiling when he extracts himself from the frame: he’s getting closer. “I’m still not up to full power,” he says. “We can go quicker. We can do it.”
The pressure to succeed and attract funding is not as intense as in previous years—winning the AHS Sikorsky prize has led to a wealth of opportunities. The pair now give talks and have begun to attract financial backing, with Google one of the team’s sponsors. “It’s given us the freedom to expand—we have finally succeeded at making our business model sustainable. The $250,000 AHS Sikorsky prize was effectively our current budget. This will allow us to take things to the next level.”
The next level is another plane and another “unwinnable” prize: the £50,000 ($74,120) Kremer International Marathon Competition. Launched in 1988, it requires a human-powered aircraft to complete a course with two turning points “not less than 4,051 meters [2.5 miles] apart,” including take off and landing, in under an hour. “No one has come close to getting it,” says Reichert. “It’s theoretically impossible. It’s perfect.”
AeroVelo’s plan features a key conceptual shift. “The big difference to previous attempts,” says Robertson, “is that instead of a single pilot, we have multiple pilots spread along the span of the aircraft, so the flight speed goes up without a substantial increase in the power required from each pilot. There’s a lot of difficulty in this solution, certainly in terms of how flexible the aircraft will be and having that many fit pilots—but it’s a simple but novel approach. Again, people say it’s impossible, but if you say that you really should qualify it. Maybe it isn’t possible in a single piloted aircraft with current technology, but our preliminary calculations show that it can be done. When people say it’s not possible, it’s because they haven’t thought about how to make it possible. And that’s our business. Making the impossible happen is what we do.”
Saturday comes around, the day of AeroVelo’s attempt at breaking the speed record. The team drives out to the track before dawn. The three crows stand by the roadside, greeting the Sun with a cacophony of squawks. Then, in a flash, a coyote comes out of the bush and takes one by the throat. It’s not yet fully light and the day has its first casualty. Will it be the only one? Will those spokes hold?
Three hours later all questions have been answered. After Reichert’s near-disastrous wobble, the team gather around a long table at the civic centre. Three front-wheel spokes are broken and it is clear AeroVelo is facing a mission-ending problem. Robertson, wearing a faded blue boiler suit dotted with Manchester United and Nasa badges, sketches out the problem, as engineers do. Reichert, still wearing sweat-stained racing Lycra, addresses the team. For once, his indomitable confidence is dented: “It doesn’t look good, but I don’t want to stop because there was an idea that we didn’t look at properly. So what have we got?”Team members throw solutions at Reichert. Each, in turn, is shot down. “Find new rims?” The nearest bike shops are in Reno, more than 200 miles away. “Build a new disc wheel out of carbon fibre?” There isn’t enough time. “Borrow some other wheels?” None of the other teams has any that will fit Eta.
“What if we just rig up the wheel with spokes and give it one more go?” Eventually Reichert submits to reality. “There are a lot of problems with these solutions. If everything goes perfectly we might inch out the record, but I might crash and die.” Gravely, he looks around the table: “It’s not worth it.”
Despite the disappointment, all is not lost: the AeroVelo bid was always about more than just breaking a record. “It’s about sharing the information and putting it out there,” says Reichert. “The ornithopter, the helicopter, the bike—none of them are going to be practical forms of transportation, it’s just not what they’re for. But the technology in this bike means it could cross North America on less than a quarter of a tank of petrol if it had a small engine. It’s not just 10 percent more efficient. It is more than that.”
With such a mission, relationships with headline sponsors such as Google to nurse and a new flight prize to chase, will AeroVelo be coming to Battle Mountain next year? The answer is a resounding yes.
“This bike is going to be the fastest next year. We haven’t even hit full power yet. This … ” Reichert says,
the fire back in his eyes, ” … is way better than the helicopter.”
Michael Hodges is an award-winning author and journalist. He wrote about Global Witness in WIRED UK issue 08.14. This article was republished from the February 2015 issue of WIRED magazine.
Nine days after leaving San Francisco, a blue car packed with tech from a company you’ve probably never heard of rolled into New York City after crossing 15 states and 3,400 miles to make history. The car did 99 percent of the driving on its own, yielding to the carbon-based life form behind the wheel only when it was time to leave the highway and hit city streets.
This amazing feat, by the automotive supplier Delphi, underscores the great LEAPs this technology has taken in recent years, and just how close it is to becoming a part of our lives. Yes, many regulatory and legislative questions must be answered, and it remains to be seen whether consumers are ready to cede control of their cars, but the hardware is, without doubt, up to the task.
What’s remarkable isn’t the fact Delphi completed this trip, but the fact several companies could have done it. Google, Audi, or Mercedes would have had little trouble handling this level of autonomous highway driving. The news here isn’t that this was possible, but that it was so easy.
“The technology is not what is most notable from this trip,” says Jeff Miller, an associate professor at the University of Southern California who works on autonomous driving. “The fact that they drove as far as they did and had a lot of publicity will help the technology more than any programming or hardware on that vehicle.”
The speed with which the technology has reached this point is stunning. Just 11 years ago at the 2004 Darpa Grand Challenge, the most advanced autonomous vehicles of the day attempted to complete a 150-mile course. The best any of them could do was 7.32 miles—and that vehicle got stuck and caught fire. The next year, five vehicles completed a 132-mile course, but took seven hours to do it. Autonomous vehicles have made enormous strides since then, which is especially remarkable when you realize the auto industry typically spends five to seven years developing a new car.
Today, most of the world’s major automakers are working on autonomous technology, with Audi, Mercedes-Benz, Nissan, and Volvo leading the pack. Google may be more advanced than anyone: The tech giant says it’s self-driving cars are so far along, they can recognize and respond to hand signals from a cop directing traffic.
Most automakers are taking a slow and steady approach to the technology and plan to roll it out over time. Most expect to have cars capable of handling themselves in stop and go traffic and on the highway within three to five years. Cars capable of navigating more complex urban environments will follow in the years beyond that, while fully autonomous vehicles are expected to be commonplace by 2040.
Propelling Us Toward the Day Humans No Longer Hold the Wheel
Companies like Google, which has racked up more than 700,000 miles with its autonomous vehicles, and Audi, which recently completed a road trip from Silicon Valley to Las Vegas, get all the love when it comes to robo-cars. But Delphi is doing just as much work behind the scenes, propelling us toward the day when humans no longer hold the wheel.
One of the auto industry’s biggest suppliers, Delphi has a solid record of innovation, from the first electric starter (1911), to the first in-dash car radio (1936), to the first integrated radio-navigation system (1994). For the past 15 years, it’s been working on active safety features (think active lane keeping and blind spot monitoring). Lately, it has been consolidating all this hardware into a holistic system that lets the car handle itself.
Delphi installed it all in a 2014 Audi SQ5, which Delphi engineers chose simply because they think it’s cool. Seriously. It has windshield-mounted camera spot lane lines, road signs, and traffic lights (in color). Midrange radars that see 80 meters sit on each corner. There’s another radar at the front, and a sixth at the back, plus two long-range units on the front and back. The front corners have built-in LIDaR.
The cross-country trip was meant to generate some publicity, yes, but Delphi also wanted to expose the system to variable real-world conditions and collect terabytes of data to further refine the technology. This car was built within the past year, but it takes advantage of tools that have been in the works for at least 15 years.
“It was time to put it on the road and see how it performed,” says Delphi CTO Jeff Owens. “It was just tremendous.”
The Delphi caravan (the self-driving car, a follow car with more personnel, and a Winnebago full of PR, photo, and video folks) followed a southern route, largely to avoid snow. Apart from the shock of realizing just how long it takes to drive across Texas, the biggest scare of the trip came while crossing a double-decker steel bridge on the drive from Philadelphia to New York. “I saw that bridge coming, and I thought, ‘Oh my gosh, this is gonna be a grab the wheel moment,’” says Katherine Winter, a Delphi software engineer. That’s because being surrounded by metal plays hell on radar by making it difficult to discern what’s a threat and what isn’t. But Delphi’s refined how its software understands the radar data and uses the other sensors to augment it. “It actually outperformed what we thought it would do,” Winter says.
Building the car helped Delphi hone the hardware and software automakers will want and need as they begin producing autonomous vehicles, and test it in a variety of situations. That included rain, hot weather, construction zones, and tunnels. “It didn’t miss a lick,” Owens says.
The team celebrated the arrival in New York with high-fives, but Delphi’s not surprised by the accomplishment. It knew before setting out it could handle the miles. It just needed to show us it could.
The six engineers who cycled through the driver’s seat only took control of the car when it encountered a situation they weren’t confident of handling safely, like a construction zone with zig-zagging lane lines, or to make an aggressive lane change to get around a cop car on the shoulder. They obeyed the speed limit and avoided night driving.
There’s no indication that it’s capable of handling the road with far more skill than a human. You’d have to look twice to spot the cameras and LIDaR around the car; the radars are hidden behind plastic body panels. Even the trunk looks ordinary, which is quite a feat—Delphi packed all the necessary computers in the spare tire compartment. That was intentional, Owens says. “We were kind of going for the remarkably unremarkable look.” The reason for this modesty is any tech Delphi pitches to automakers has to be unobtrusive and production-ready.
That is the ultimate goal here. This car won’t be in showrooms. But the stuff that makes it work certainly will be. Delphi makes all the stuff automakers don’t (or can’t) make themselves. The plan is to offer everything an automaker might need to make a fully autonomous car. It’s an off-the-shelf solution anyone can use.
“This drive is one more marker on the exciting road toward automated vehicles,” says Bryant Walker Smith, an assistant professor at the University of South Carolina School of Law and affiliate scholar at the Center for Internet and Society. He studies autonomous vehicles and says Delphi’s accomplishment raises public awareness “by previewing what will someday be possible.” That’s a good thing, as long as the conversation includes “what was required, what was hard, and what remains to be done.”
Delphi will take a few weeks to dissect and digest all the data it gathered and everything the engineers noticed, like the car’s skittishness around tractor trailers, and adjust the system as needed. Then it might be time for a trip through Europe, where Delphi does a lot of business and automakers are keen on both active safety and autonomous features.
For now, though, the company is pleased with the progress it’s made, and it confident it will play a significant role in the coming shift to self-driving cars, Owens says. “Delphi can march at the same speed as Silicon Valley.”
Karl Marx was supposed to be dead and buried. With the collapse of the Soviet Union and China’s Great Leap Forward into capitalism, communism faded into the quaint backdrop of James Bond movies or the deviant mantra of Kim Jong Un. The class conflict that Marx believed determined the course of history seemed to melt away in a prosperous era of free trade and free enterprise. The far-reaching power of globalization, linking the most remote corners of the planet in lucrative bonds of finance, outsourcing and “borderless” manufacturing, offered everybody from Silicon Valley tech gurus to Chinese farm girls ample opportunities to get rich. Asia in the latter decades of the 20th century witnessed perhaps the most remarkable record of poverty alleviation in human history — all thanks to the very capitalist tools of trade, entrepreneurship and foreign investment. Capitalism appeared to be fulfilling its promise — to uplift everyone to new heights…