A bad algorithm: why champion racer lost at Monaco

  • 27-May-2015 02:18 EDT
f1moc2015_jk1711174.jpg

Racing strategy decisions are made based on data generated by about 200 sensors on modern Formula One cars (like this one piloted by Lewis Hamilton at the 2015 Monaco Grand Prix) and its wireless transmission to the pits and remote centers.

Lewis Hamilton, the reigning Formula One world champion, dominated the Monaco Grand Prix on May 24, performing flawlessly for 63 of the race’s 78 laps. His ultimate third-place position was not due to subsequent driver error, but instead a lapse in data transmission and its interpretation by the Mercedes AMG Petronas technical team. The crushing result for Hamilton reveals the critical role that sensor-generated data coming off Formula One racecars plays in the world’s premiere motorsport.

Viewers deafened by the roar of engines might not be aware of the silent crunching of data generated by about 200 sensors on Formula One cars—and its wireless transmission to the pits and remote centers. “We have to follow the data. This is how the sport works,” said Toto Wolff, Head of Mercedes Motorsports. “You try to get as much input as possible from the engineers, from the management, from the driver, and then take a decision. The algorithm was wrong.”

Hamilton had a decisive 25-s lead on lap 64, when 17-year-old rookie driver Max Verstappen of the Toro Rosso team slammed hard into a Lotus then the barrier at the Sainte Devote first corner while battling for 10th place. When cars were slowed down for safety, the Mercedes team called Hamilton in for what turned out to be an unnecessary pit stop. New tires were dashed on, presumably to allow Hamilton to claim an easy victory. However, when he returned from the pit to the track a few seconds later, Mercedes teammate Nico Rosberg and Ferrari’s Sebastian Vettel had pulled ahead—leaving Hamilton in a disappointing third place where he remained until the finish. “We thought we had a gap, which we didn’t have,” said Wolff.

Fast connections

The day before the Grand Prix, Mercedes and wireless telecommunications company Qualcomm announced an agreement in Monaco to extend their existing technology partnership. “We do everything we can do to make the car go fast,” said Thomas Weber, a chief of technology at Daimler, at the event. “Connectivity plays a major role.”

Daimler, the maker of Mercedes cars, has been collaborating with Qualcomm for about 10 years. “The car is generating all kinds of data from many different systems,” said Derek Aberle, Qualcomm’s President. “They have a limited amount of time on the track, so the quicker it comes off the vehicle, the quicker they can analyze it.”

The course in Monaco is well known for its narrow path and hairpin turns, but it is equally treacherous for communications. The pit blunder was mostly attributed to a botched judgment call, but was also partly blamed on a freeze in data communications at a critical moment, which could have compounded or caused the miscalculation.

Monaco is a dense urban location, providing uncertain GPS and connectivity—as evident by recurring blips in the big screens serving up video and audio to the massive crowd assembled around the course.

“The ability for a driver to use advanced wireless systems that effectively communicate is a real challenge,” Aberle told SAE Magazines. “There are a lot of different things interfering with one another.”

He explained that Qualcomm is exploring how to blend cellular, Wi-Fi, and close-range 60-GHz technology—each with its own benefits and limitations—for greater overall speed and reliability. The Mercedes team did not identify the exact role of Qualcomm’s technology, or any specific data glitch, in the Grand Prix race.

Rather than rely on real-time transmission of data over the two-mile course, data are commonly first collected by the racecar’s computer. When the car is within range of the pit, the data are transmitted by a small radio antenna on the rearview mirror. “The larger the area we can give them for reception, the more data can come off the car,” said Paul Jacobs, Chairman of the Board, Qualcomm.

It’s a technology race

According to Deiter Zetsche, head of Mercedes cars, there is also a team of about 30 engineers located in Great Britain receiving and analyzing data. “They are doing something like war games, asking ‘what if’ questions,” he said. Zetsche explained that in a recent F1 race, the Mercedes team changed its strategy at the last moment. “They asked the driver to pit, and he won the race.” The split-second call for a pit stop in Monaco had the opposite effect.

Qualcomm executives explained that technology lessons from Formula One could soon enable manufacturers to remotely monitor engine health of your daily commuter car, or even your own personal health. In fact, among the data sent to the pits from a Formula One racecar is the driver’s heartbeat.

“We’re exploring how much machine intelligence can go into the car,” said Jacobs.

He said everyday cars will “become a friend in a way.” He also equated connected cars to wearable technology. “I strap it around my body, and it augments my human physical capabilities, and going into the future, your mental capabilities as well,” he said.

Yet, as evidenced by the blunder that cost Lewis Hamilton the race, the balance between engine power, computer power, and brain power can be delicate, especially when the difference in lap speeds between competing Formula One cars are measured in the thousandths of a second. “We got the math wrong,” said Wolff. “The calculation was simply wrong. That’s what happened.”

Share
HTML for Linking to Page
Page URL
Grade
Rate It
4.00 Avg. Rating

Read More Articles On

2016-11-13
Lengthy automotive development and production cycles have long prevented automakers and startups from working together. While that’s changed a bit, many young companies still find it difficult to work with OEMs.
2016-11-13
Focused on the near-term safety-improvement potential underlying autonomous-driving technology, Toyota - counter to much of the auto industry - sees real promise in developing SAE Level 2-3 systems.
2016-11-15
Tanktwo, a Finland-based startup company is rethinking the basic battery cell and challenging the fundamental economics and operational assumptions of EVs. The ingenious concept is worth engineers' attention.
2016-11-14
Conti’s 48-V system will be standard equipment on both gasoline and diesel versions of the Scenic Hybrid Assist model. It is the first of multiple 48-V production announcements coming over the next few years.

Related Items

Technical Paper / Journal Article
2010-10-25
Training / Education
2007-03-01