Formula One brings its high-tech, upscale racing circus to Las Vegas for the city’s inaugural Grand Prix this week, capped off by next Saturday night 200-mile-plus per hour, 190-mile jaunt around the streets of the gambling mecca, including a straightaway down the world famous Strip.
The globe-trotting F1 is in Vegas for the first time in more than 40 years. So why will the 20 drivers as they descend into their cockpits know the recently built road course like they have raced it before, from where to slow, what type of tires to deploy, when best to pit stop for a wheels change, the best locations to overtake, and how to handle the 17 corners and bends? Ask any driver; Red Bull’s Max Verstappen or Mercedes’ Lewis Hamilton, to name the best known competitors, they will have “driven” the course scores of times.
The answer lies in the big money competition between teams to gain a technological edge through data; and a scramble in the days leading up to the race to transmit car and track sensory reading back to their home base technology hubs to glean the slightest insight or advantage.
“Ahead of a race weekend, we'll run about 250 million simulations of the race that's using our on premise hardware here at the Technology Center (in England),” said Edward Green, Head of Commercial Technology at McLaren Racing. McLaren’s two F1 drivers, Lando Norris and Oscar Piastri, will have spent days in a physical simulator in the United Kingdom in the week leading up to the race.
In vivid detail, including audio, the simulators mimic the car racing on the course. The machines are programmed using data collected from sensors on the cars–McLaren has 300 such devices on each of its two entries. “Air pressure sensors, tires, engine, will have sensors that will scan the surface of the track underneath the car,” Green said.
The data is fed into a mobile data center at the track, then relayed back to England, where McLaren works with Dell and Google Cloud for supercomputing. That information is fed into the simulator, which is a cockpit on a moving base surrounded by video screens, creating a hyper-realistic virtual reality of the course and speeding car. This is the standard practice for the 20 teams.
But this approach largely only works for races where there is historical data, and where there is an existing LIDAR (Light Detection and Ranging), or laser scan of the course. That’s obviously not the case in Sin City. In the case of Las Vegas, months ago the city allowed a firm to scan parts of the course. Combined with a Computer Aided Design (CAD) model, this allowed teams to have a Vegas simulation, but it is clearly incomplete.
So once the roads were closed and the track recently completed–-F1 is spending $400 million on building out the temporary venue—a more comprehensive LIDAR scan would have been taken and shared with the teams.
“This means that from the week before the Grand Prix, the teams have a significantly better model and therefore better simulation output results,” said Rob Smedley, a former engineer for three F1 teams. “It's an imperfect world for the teams in new tracks like this because all of this work usually gets done at least three months out from the race, whereas in this case it is a matter of days leading up to the Grand Prix before the teams can start a detailed simulation.”
By that time most drivers are onsite in Las Vegas. But each team has simulator drivers who will spend hours in the machine, and then relay their findings to their onsite peers. The simulators not only give the drivers a sneak peek at the track, it informs the engineering team of changes that might need to be made to the car.
“The engineers and drivers arrive at the track with a car that is already 90% optimized (meaning it) has the correct set up for that particular” track, Smedley said. “Then they will have an array of options depending (on how) the car balances when they test on the actual track. These options will have been derived previously from the simulation tools.”
The drivers get practice sessions, as well as the qualifying race that determines their position at the start (known as the grid). Data from these runs is relayed to the teams’ home bases and fed into the simulator. After the qualifying run, no changes are allowed to the car according to F1 rules, so teams are rushing in the days before a race to get data to their simulators that then might dictate vehicle changes or driver strategy.
McLaren’s Green said in fact his team developed new car parts to account for Vegas’ lengthy straight away down the Strip, a feature that can spark faster speeds.
“'I'm not allowed to share” what the part is, he said. “But obviously, because it's such a low drag (meaning high speed) down the back of the Strip, there's some obvious parts that typically tend to change. It’s unusual to be sort of developing new parts for the car this late in the season.”
Of course, data and advance planning carry teams only so far. Earlier this year F1 had a first time race in Qatar (laser scans would have been available well in advance as it is contested on a permanent track). Despite the night time scheduling, the heat was brutal.
“And they actually discovered that the tire wear because of the heat and because of the characteristics of the circuit, they had to limit the amount of time that the drivers were able to use a single set of tires so no tire could be used for more than 19 laps,” said Neil Ralph, Principal Sports Partnership Manager, Amazon Web Services, which is F1’s official data company. That mooted teams’ painstakingly developed strategies of when to pit stop to change tires, which is a critical element of F1.
Heat certainly should not be an issue in the desert at night in November in Las Vegas. But what else could lurk out there that the F1 teams have not considered? If sports is the theater of the unexpected, we’ll see this weekend if the computer models and sensors were enough to account for all the possibilities or if Vegas delivers a wild card.