Digital Twins: Revolutionizing The Development Of Formula 1 Cars With AI And Cloud
In recent years, the concept of “digital twins” has gained significant attention across various industries, presenting new opportunities for companies to enhance their products and streamline costs. By integrating data analytics, machine learning, and the Internet of Things (IoT), digital twins enable organizations to improve decision-making, minimize downtime, and boost efficiency.
F1 engineers aim to design a car that feels like it’s on rails on the track and can achieve high speeds in corners while maintaining aerodynamic efficiency to avoid slowing down on straight stretches. With the introduction of time restrictions on wind tunnel and CFD usage in 2021, virtual design has become even more crucial for performance.
What Are Digital Twins?
Digital twins are virtual images of actual processes or objects that enable the simulation and real-time observation of their behavior. These models, created using data from sensors, IoT devices, and other sources, can be used to evaluate and enhance the performance of physical entities. Digital twins facilitate scaled testing, shorten production times, and improve products and services across various industries, including manufacturing, healthcare, and transportation.
First introduced by Michael Grieves in 2002, the concept of digital twins has evolved with advancements in machine learning algorithms and IoT technologies, now offering predictive analytics and real-time insights.
In manufacturing, digital twins optimize and reproduce production processes, reduce downtimes, and improve product quality. Airbus, for example, uses digital twins to test and simulate aircraft production processes to identify problems and streamline the assembly line. This approach has improved product quality while reducing production costs and time.
The digital twin concept consists of three parts: the physical object or process and its physical environment, the digital representation of the object or process, and the communication channel between the physical and virtual representations:
source: The three key aspects of the digital twin technology
F1 And Digital Twin Applications
Drivers can train and learn in a car simulator (a digital twin) before taking to the track. Since testing time is limited to save money, drivers rely heavily on these digital versions of the car and track. The more accurate the simulator is, the better the drivers can prepare in practice.
In Formula One racing, digital twins play a crucial role. While the team vehicle races at 300 km/h on the circuit, a crew of technicians and engineers monitor the car’s stresses in real time from the pit, making minor adjustments to ensure peak performance. Digital twins are used to simulate the car’s performance before heading to the racetrack, allowing for the construction of the best prototypes. This reduces the design time for each iteration, leading to improved product development.
This is an old dashboard of the McLaren Mercedes team:
source: digital twins and the use to simulate the performance of the cars
Digital Twins: a key technology in F1 racing
Contrary to the belief that F1 drivers know every track and need little preparation between races, they engage in extensive preparation before their first practice session on a track. This work is conducted at the team’s headquarters and training center, employing advanced technology similar to that used during the actual race.
The Fédération Internationale de l’Automobile (FIA), the governing body of Formula One, has strict regulations regarding on-track car testing during the season. Each F1 team has a driver training facility with dedicated simulations to understand their cars better.
Simulators provide real-time feedback on a car’s performance, allowing drivers to make adjustments according to course requirements. They can also simulate various scenarios, such as day/night cycles and different weather conditions, helping drivers understand the physical demands of a track.
In the next picture, Monolith AI enables F1 Teams to monetize their CAD, simulation, and test data, using it to predict the behavior of designs without performing new simulations or physical tests:
source: Artificial intelligence for test labs
Benefits achieved by Formula 1 relative to the Digital Twin concept
- Reduced cost of development, testing, and certification using AI/ML and simulation.
- Applying the technology to power other verticals into the Team.
- Working with real-time data and analytics.
- Empower engineers when making difficult decisions.
The F1 Teams and Digital Twins
In the last Formula One season, the budget ceilings of the racing teams were reduced from 145 million dollars to 140 million dollars per team. This cut has forced individual teams to focus more than ever on cost efficiency and resource management. Technology has become more important to Formula One racing teams than in previous seasons.
Epicor Formula 1 partnership with Visa Cash App RB
Scuderia Visa Cash App RB partnered with Epicor to focus on continuous performance improvements both on and off the track. Epicor Kinetic, designed for manufacturing, serves as the data management engine at the core of the team’s operations, streamlining production, linking components to telemetry data, and maximizing car performance. It automates job creation processes, reduces data input, and increases productivity.
Epicor Kinetic also provides detailed visibility into the machine, operator, and tools used to manufacture each piece, developing cost analysis models to inform purchasing decisions and quickly correct issues before the team hits the track. An update service automates software updates, minimizing system downtime and ensuring smooth transitions to new versions.
Detailed insight is provided into the machine, the operator, and the tooling used to produce each part. This enables the development of cost analysis models to inform purchasing decisions and allow for quick correction of issues before the team hits the track.
For a sport that is constantly evolving, Epicor has developed an update service to automate software updates and minimize system downtime. This will allow the team to smoothly transition to the latest version of Epicor Kinetic this year without interrupting business operations or requiring additional training for operators, freeing up more time for the track.
Epicor Kinetic enables the VisaCash App RB to automatically create a complete build plan of a race car, tracking each part in sequence and storing a snapshot that encodes the structure for a specific race. In addition, Epicor Kinetic also automates the order creation process for each part, reducing the amount of data entry required and increasing productivity.
In the next image, we have the different modules and applications of Epicor Kinetic, like manufacturing and distribution, quoting and sales, pre-production materials planning, inventory management, etc:
source: Epicor Kinetic Enterprise Resource Planning system
Red Bull Sim Racing and F1
Oracle Red Bull Racing’s official e-sports squad will use the same technical and strategic methods as its F1 outfit. The Sim Racing team uses cloud-based computing infrastructure, real-time data dashboards, and lap-by-lap analytics of braking, acceleration, and throttle to assist drivers in improving their performance in virtual races.
Oracle Red Bull Sim Racing will also get the opportunity to demonstrate its amazing technical skills and, in an unprecedented move, share driver data with fans. Historically, F1 teams have guarded their data rigorously to prevent disclosing crucial information to opponents. Fans rarely see their favorite driver’s specific performance or telemetry data.
The relevance of computerized simulations like this is rising across industries, not just sports. Oracle Red Bull Sim Racing’s experience absorbing, sharing, and analyzing data from its digital simulations is the most recent illustration of how real-time data monitoring and quick analysis may provide new opportunities. Digital innovators in areas such as manufacturing and logistics are increasingly relying on “digital twins,” which allow businesses to test equipment stress limits or experiment with alternative plant or warehouse arrangements without incurring significant costs or investments.
In the next image, they can be sped up and run in parallel with other cars and teams to view the vehicle dynamics and the strategy groups:
Source: A look at the full race simulations
Summary
At the factory, actual data and observations validate the digital twins. Each car streams 100,000 data points per second, generating over 100 GB of data every race. Incorporating virtual sensing increases real-time data, making it essential for digital twin use in F1. McLaren uses Apache Kafka to manage the car’s data, providing a high-throughput, low-latency platform for real-time data feeds.
Three primary information sources — pitwall and course cameras, driver awareness, and sensory data from sensors — identify car issues. Sensory data fed into a digital twin can predict component degradation before failure. Extensive testing and comparing simulation data to real-world results help achieve a reliable vehicle.
In Formula One, split-second decisions can determine the race outcome. Teams rely on digital twins to run scenarios and prepare for unexpected developments, though surprises can still occur.
