Location Intelligence

and geospatial in F1

The pinnacle of motor sport is Formula 1, drivers race around tracks in million-pound cars designed to be as efficient as possible at speeds over 200mph. The cars are demonstrations of innovation, with most of the technological breakthroughs filtering down to modern day cars. Without F1, we may never have had antilock brakes (ABS), carbon fibre components, adaptive suspension and even the buttons around the steering wheel.

As the cars have become more refined, mere tenths of a second can mean winning or losing a race, so measuring real time performance and position is vital, and all this technology also gets filtered down for use by business.

They fit each car with over 120 sensors, which generate over 1.1 million telemetry data points per second transmitted from the cars to the pits during a race. These monitor everything from the car’s speed and acceleration to the steering angle to the charge in the energy recovery system to the mass of fuel used to the status of its drag reduction system. This data has to be used in near real-time to solve many of the challenges which occur.

Since 2018 AWS (Amazon Web Services) has been providing the graphics and real-time analytics for F1, showing information on-screen about pit stop strategies, overtaking opportunities introducing information on the ‘undercut’ and ‘overcut’, this is where an F1 driver gets an advantage by either taking a pit stop early or later. These graphics are underpinned by clever real-time models which allow viewers to see when a driver is at threat of losing their position due to another driver pitting or not pitting. Since we know how long a good pit stop may take at a particular circuit and we also know the speed advantage of driving in clean air and the performance improvement of the new tyres, these factors can be put into a model (formula) which will take the times for the 2 battling drivers and relate the difference in their projected times after the pit stop has occurred. AWS shares this model as:

To the viewer, this graphic provides an exciting insight which adds an extra element to the race, but it isn’t magic or super computer math. This is quite simply well-structured data used in a model to provide a clean, clear interface, much like many businesses may need to do when running logistics or construction material removal. Factors could be substituted from tyre degradation to carbon footprint. The only difference between the system AWS uses here and the common cloud computing used is that AWS has turbo charged their throughput to be able to run more calculations faster.

As we previously discussed, when you have cars zipping around a circuit at 200+mph, you need to have a fast and responsive system. From the moment the signal on the car is sent to the time it is ready to be used by the race director and TV is less than 500 milliseconds. Once a signal is captured at the race track, it begins its journey, first passing through F1 infrastructure to an HTTP call to the AWS Cloud. AWS and F1 used Amazon API Gateway to act as the entry point to the application, which was itself hosted as a function in AWS Lambda to implement the race logic. Once the function received the incoming message, it would first update the race state stored in Amazon DynamoDB (for example, change of driver position). If the function determined that it was a trigger for a prediction, it would use the model trained in Amazon SageMaker to make and return the prediction as a response to the call, ingested back through F1 infrastructure to the broadcasting center.

How can a driver turn up to a race they have never raced and be quick on day one or how can the engineers know how the car will perform? Quite simply they drive the car around a giant digital twin.

The tracks and surroundings are captured using high resolution LiDAR, ensuring that every bump and visual cue is in the system, this is vital in ensuring the full immersion and accuracy as a driver will complain if they can’t feel the bumps in the track at Monaco or if the turn 3 is 2cm too wide at Silverstone. Teams like Red Bull Racing and Mercedes work closely with gaming companies to ensure that their simulators are as realistic as possible. To put this into context, a new circuit which we’d planned to capture, was planned to use some Leica Pegasus Two Ultimate mobile scanners to drive the circuit, each capturing around 1 million points per second.

These simulators are called ‘Driver in Loop’ (DiL) simulators and as well as the track digital twin that is shown on screens in front and around the driver there is also a cockpit which the driver sits in (sometimes called a tub), which sits on actuators that move perfectly matched to the drivers input.

The simulators are so accurate that different racing setups, weather and even tyres may be used in preparation to race data so that the driver can work on their set up in advance of the race. The simulator takes vast amount of cost away from the mechanical work which may need to be done to get the balance right as the set up can be altered at the touch of a button. Also, technicians can get real-time data back on performance and expected deltas as wells as show playback of areas of improvement.

In the run up to the Silverstone 2021 Grand Prix, Lewis Hamilton announced on the morning of the race that he had finished qualifying the previous day and had gone immediately to the simulator to do a few hours practice, which from what Mercedes has said about the regime they have before a race weekend, the drivers may practice in the simulator for that track for almost 2 days solidly.

The simulators have proven essential to the F1 teams, so much so that every team has one for every driver and test driver. It also provides information to the technicians ahead of a race which may help with strategies and choices. Much of this has begun moving to common business, maybe not in the guise of a race simulator but in the form of highly detailed digital twins which may be tested for real world results. When we look at the automotive industry, they now use immersion rooms which use digital twins of the cars planned to be built.

Surprisingly, it isn’t the race itself where the majority of geospatial technology is used, it is in the logistics of transporting the cars, drivers, components and offices to and from the race tracks across the world every couple of weeks.

Logistics providers make extensive use of GIS data, using it to plan delivery routes, for example, and predict arrival times based on weather conditions, congestion, and known delays at ports, airports, and border crossings. Digital twins of will also help providers to optimize their conventional logistics networks, for example by using rich data on customer locations, demand patterns, and travel times to plan distribution routes and inventory storage locations.

When every fraction of a second counts and there are millions of pounds at stake only those who innovate and develop cutting edge solutions will win the season. As mentioned earlier the cutting-edge innovations filter down to everyday business.

If we look at the super-fast blink of an eye response times which are being used by AWS for the graphics and analysis, they are available now, fully ready to be implemented in your business and could be used for many different applications, from logistics through to fleet management and even mapping of IoT sensors. When looking into the AWS solution, there is a wealth of information and is almost ready to go out of the box, there are even prebuilt machine learning and artificial intelligence models which can be implemented.

Although the simulators may not be something every business requires, the way which they are used very much is. Formula 1 has shown the importance of digital twins, from the building of the car through to the circuit and logistics, in formula 1 everything has a digital twin which can be tested to its extremes and more importantly works in harmony with one another.

Like many engineering companies, Formula 1 has realized the gains to be had with digital twins, though it is the extension to other areas of business which has provided the grand prix wins, although the F1 industry use the high-end equipment, that doesn’t mean that this is out of reach for your average business, it is just a case of scaling down to sensible levels, scanning to millimetre accuracy is nice, but is it necessary!