Smart cycling: how IoT and edge computing power the Tour de France

For many years, crowds jostled for space along the 3,000km-plus Tour de France route to catch a fleeting glimpse of the peloton speeding past. Those roadside viewers had no way of monitoring individual riders or, say, predicting who might wear the yellow jersey, and fans the world over had to rely on radio and TV broadcasts for race updates.

Now, in the age of connectivity and IoT, the world’s best cyclists also race along a range of advanced digital pathways that are closely monitored by officials and fans, in real time.

As the Official Technology Partner of Amaury Sport Organisation (A.S.O.), the Tour de France organizers, NTT DATA is responsible for building and managing those pathways.

We use our edge-computing and cloud technologies to create the world’s largest connected stadium by setting up a digital twin of the race, which combines layers of real-time information to replicate all aspects of this highly dynamic event digitally.

In this way, we bring millions of data points to life through insightful visualizations across a range of platforms, serving information to fans, race officials and broadcasters.

From the fans’ perspective, we’re giving them a more detailed understanding of what's happening in the race – for example, with data-driven insights and AI predictions on our LeTourData channels on Twitter, Instagram and television. For A.S.O., we help their operations teams to communicate and plan more efficiently in less time than before.

Bicycles on the edge

At the core of our data-gathering are the trackers installed on each bicycle. They transmit a constant stream of latitude, longitude and speed data over radio networks to race motorcycles or a plane. From there, a microwave signal carries the data to the end of the race, where a truck-based edge-computing device runs a containerized version of our real-time analytics platform.

As the data arrives, it is fed into algorithms that calculate a range of race insights such as rider groupings, their location along the route and the distance between groups.

The locally processed data is then sent to a custom application that gives A.S.O. officials a detailed, real-time view of what's going on in the race without having to rely on mobile coverage or other modes of data transmission. The application runs on Surface Pro X devices, which have enough battery power to outlast a race stage.

We place other trackers in the official race vehicles, and these are connected via Transatel SIMs to give accurate locations for the vehicles. They have also been installed in the Shimano neutral-service vehicles, which provide mechanical help to riders who have been isolated from their team cars.

So, in addition to giving the race organizers a real-time view of the riders and the race environment, we’ve added a layer that shows the locations of the official vehicles.

Until about two years ago, officials had to rely on radio communications to do their jobs, and now they have this level of detail on demand. It is a great example of how IoT devices and edge computing have brought a new level of insight and efficiency to the Tour de France.

• ALSO READ → Detailed data delivers a fan-first experience at the Tour de France

WATCH | How we use edge-computing and cloud technologies to create the world’s largest connected stadium by setting up a digital twin of the Tour de France

Fleshing out the digital twin

While our edge-computing device is responsible for analyzing and transmitting real-time race data locally, the raw data is also sent to the Microsoft Azure cloud computing platform, where another layer of data analysis generates a suite of processed data to be made available via an application programming interface.

Apart from this data, we also use other types of information to put together the digital twin, and they don’t all come from IoT sensors.

We view the race environment in three distinct ways: the actual race and the riders, the organization or event (the people who work there, their vehicles and their movements) and the fans.

• The race: There is widespread tracking of riders and race vehicles, and the data is processed locally in real time, as explained above.
• The event: Here, we bring together the start and end location of each stage, along with the race villages, the VIP zones, the technology zones and much more, to create a static map.
  
  Then we add a dynamic layer by integrating with fleet management systems for the various race trucks to position them on the map. We also deploy other IoT sensors to give us data on race-village entrances, exits and other useful areas – even the toilets. It’s a real time-saver not to have to hunt down our NTT Tech Truck at the end of every stage! There’s also a feed of weather data.
  
  Our truck is a central workspace that we share with A.S.O.’s technical team. From here, we can assist with the network infrastructure we deploy along the race route. If things go wrong, we can just walk to somebody else’s truck and work with them directly instead of remotely.
• The fans: We use technology to gather information about the crowds along the route. For example, people’s mobile phones broadcast constantly, even when they’re not connected to a wireless access point. We don’t know who the fan is, but we know somebody's there and we can use that data to count the number of people in a crowd. This year, we’ll also deploy AI-powered video-vision technology to assist in this regard.

WATCH | The Tour de France data journey: where it starts and where it ends

Another integration is with the Tour de France accreditation system, which monitors race officials and other accredited people tapping in or out at various locations.

All of this data comes together – across four different networks to avoid congestion-related delays – to give us a deep understanding of the volume and movements of people in different areas.

This information is also sent to our Microsoft Azure-based analytics hub to feed into the digital twin. We plug two downstream applications into that data model: a Tour de France operations application and an event app provided by A.S.O. to people working on the race.

• Download and read our 2023 Tour de France technical white paper

Meet Marianne, our digital human

Also making a return this year is our digital human, Marianne, a lifelike human avatar developed by NTT DATA and presented on a kiosk equipped with a large, high-quality screen and a camera. The avatar is named after Marianne Martin, the US cyclist who won the Tour de France Féminin, an early version of the women’s race, in 1984. 

She can interact with humans through voice and visual prompts to share Tour de France information, both from a broad database about the race and from an integration with our live race data. So, Marianne can tell fans about the history of the Tour de France or explain what the yellow jersey is, but also say who is leading the day’s race stage.

This is an exciting new category of technology with many applications in sport, retail and beyond: think of virtual shop assistants answering customers’ queries, or information booths in the tourism industry staffed by smart, interactive virtual guides.

This year, Marianne will also include a ChatGPT integration to enhance her capabilities and broaden her scope of reference. The kiosks will be installed in the NTT DATA Tech Truck and in the VIP areas of the race villages, although the technology is likely to become available to fans on their phones in years to come.

A fast-moving, three-dimensional puzzle

In an event as geographically dispersed and dynamic as the Tour de France, many things can – and will – go wrong.

It’s not like an IoT solution in a factory where you can quickly find, assess and fix a problem on the factory floor. Someone could accidentally kick cables somewhere in a race village and break the connection, and then you have to act quickly to understand what's going on.

During the race stages, everything in our technical ecosystem needs to be active and working simultaneously before we start preparing all over again for the next stage. It's like trying to build a three-dimensional jigsaw puzzle that's constantly moving.

A common problem is when riders swap bicycles – after a crash or mechanical failure, for example – and lose their IoT sensor. We have other ways of monitoring their progress to the end of the race, but we always have to be prepared for such an eventuality. It may not even be immediately clear to us that a rider has swapped bikes. We have algorithms that will flag an IoT device’s diversion from the route, so we can then communicate with officials to find the cause.

We prepare for these challenges as best we can by testing significantly ahead of time and making sure we have plenty of spare and backup equipment.

At NTT DATA, we pride ourselves on being prepared for anything that comes our clients’ way in terms of their networks, edge and cloud computing, and infrastructure. Offering a service that is both innovative and highly reliable is what makes us a sought-after managed service provider – and the Tour de France is where we can demonstrate this to the world.

• ALSO READ → Tour de France Femmes avec Zwift breaks gender and tech barriers

Following the announcement last year of the integration of NTT Ltd. and NTT DATA to create a USD 30 billion IT services powerhouse, our technology partnership with both the Tour de France and Tour de France Femmes avec Zwift now falls under the NTT DATA brand. Our purpose is to transform businesses for success, disrupt industries for good and shape a better society for all.