Why drivers aren’t exempt from harnessing AI’s advantages
It is evident that artificial intelligence (AI) holds transformative potential across various domains, a fact often misconceived by many.
Contrary to popular belief, mere automation does not equate to AI. True AI involves software that learns, improves, and autonomously makes decisions. Currently, in the realm of Formula E, the use of AI beyond social media applications remains limited. However, the prospect of its broader integration looms on the horizon.
Central to making AI work effectively is the availability of quality data. Given the wealth of data generated in motorsport, it stands as an ideal candidate for embracing the AI revolution. Beyond enhancing performance, there also exists a vast opportunity to leverage AI systems commercially. Superior solutions across myriad areas, from optimising electric motor coil wiring to streamlining magnetic flow or reducing weight, can result from the development of specific models.
When employed judiciously and complemented by adept software development, AI emerges as a potent tool for enhancing efficiency, reliability and safety. Its implementation facilitates the processing of vast data volumes and expedites simulations, thereby improving the cost-effectiveness of motorsport operations more widely.
The magnitude of data generated during a race weekend surpasses human capacity for comprehension and performance extrapolation. AI, however, can swiftly process car data to devise solutions that might easily elude human cognition and achieve potentially better outcomes.
Failure to embrace AI translates to falling behind the curve. Analogous to not using the internet, abstaining from AI adoption is self-defeating because it deprives people of access to crucial data reservoirs. While its adoption remains voluntary, integrating AI into racing team operations or software development emerges as a logical consequence for optimising performance and problem-solving efficiency.
The level of data that is accumulated during a Formula E weekend is beyond levels of human comprehension, which is where AI can make a difference
Photo by: Sam Bloxham / Motorsport Images
Envisaging human-AI collaboration across various domains is plausible. In the long run, a scenario in endurance racing where cars alternate between autonomous and human-driven stints, with AI learning and optimising from human input, is entirely conceivable.
PLUS: How motorsport is embracing the opportunities of AI
In the interim, we may witness real-time car set-up adjustments based on AI analysis of cornering tendencies when tyres degrade. Drivers currently do this via the steering wheel, but if the hardware they want to adjust like the diff or wings were software-controlled rather than mechanical, they could already be primed to embrace the possibilities of AI.
Furthermore, AI holds promise as a tool for drivers in pre-event simulator sessions as the technology matures. Transitioning beyond the driver-in-the-loop model entails creating simulator drivers mirroring human limitations while striving for perfection. The prospect of a realistic ‘virtual twin’, combining the qualities of the world’s top drivers, presents unprecedented opportunities for performance benchmarking and enhancement.
The commitment of significant hours to simulator sessions before every race underscores the value drivers place on preparation. But an optimised virtual twin capable of accumulating extensive simulated experience can serve as the ultimate reference point to refine on-track performance comparisons.
It’s clear to me that drivers of the future will compare data with their optimised virtual twin and learn from it. This is just a matter of time.
In light of these developments, the inevitability of AI’s ascendancy is indisputable. Those who fail to embrace its potential, including racing drivers, risk being left in the wake of progress. Motorsport has and always will be so.
di Grassi warns that drivers who don’t use AI are at risk of being left behind
Photo by: Andrew Ferraro
