Anika Begay
Formula One backer Liberty Media may have pushed the sport deeper into its American heartland: now there are Three Grand Prix in the United States, but it is Nascar that continues to increase its television audience there, compared to a slight decline for the “open-wheel” championships of F1 and IndyCar.
European racing fans are notoriously wary of stock car racing, but there’s something about a seemingly low-tech, normally aspirated, 358-cubic-inch (5.8-liter) pushrod V8 making 670 horsepower tearing around an oval and hitting the parts that other racing series can’t. Or have given up on.
Not that the NASCAR grid isn’t trying to gain a technological edge in any way possible. Lenovo is working with one of the series’ biggest names, Richard Childress Racing, to help refine its pit stops during a race, and there are a lot of them in the NASCAR Cup Series, anywhere from five to 12 depending on the track and what’s happening on the track. In particular, the company is using AI to get real-time information about fueling.
Fuel consumption is obviously a critical part of any NASCAR race, almost an art in itself, as well as a source of drama and danger. (NB: refueling has been banned in F1 since 2010 for cost and safety reasons.) The cars themselves do not have in-cockpit fuel gauges, so it is up to team strategists to constantly monitor the amount being pumped in during a pit stop and the rate at which it is being consumed.
As with any use case, fuel consumption depends on a number of variables, including the length and configuration of the track and the speeds at which the cars are running. There are a number of “cautions” during a race, at which point the cars typically consume half their fuel.
In NASCAR, drivers also use the “draft,” a technique that allows them to maintain speed in the pack without using full throttle. Less fuel consumed means fewer pit stops, and when they do, they use less. On average, a NASCAR Cup Series car, which is not the most energy efficient device, will consume 100 gallons (380 liters) of fuel in a race.
Lighter is always faster
It’s not an exact science, but Lenovo’s AI team is aiming to get as close to it as possible. If RCR could measure the amount of time gas cans were attached to its cars, they reasoned, then the team could more accurately calculate how much fuel was dispensed.
That was the brief. Lenovo’s response was to come up with a system that used in-car transponders and a camera mounted above the RCR box to identify when a car entered the box and trigger a real-time video feed.
“An AI engine looks at every frame and classifies whether the fuel can is plugged or disconnected,” explains Sachin Wani, data scientist at Lenovo AI. “We’re working at 30 frames per second, so the information is accurate to about 0.03 seconds. Before that, the gas station attendant knew he had to pump about seven seconds of fuel, without any backup device for safety reasons.”
“So it was basically a mental calculation, which meant seven seconds could become eight or nine. Or worse, five or six. That obviously throws the strategy out of whack and creates a situation where they’ve refueled and have to make another pit stop,” Wani says.
