Beyond Math integrates a physics-based AI simulation on F1 cars

The challenge of simulating the real environment at a useful level of detail is enormously complex. Vehicle and aerospace firms' design teams are being held up by traditional methods; however, Beyond Math is putting AI to work with a novel approach to world simulation that could save them from having to wait days or even weeks. "In contrast to language, where we lack mathematical models to specify the meaning of the following word, physics provides us with such models. Furthermore, machine learning is proving to be highly proficient at computation as well as pattern identification, according to co-founder Darren Garvey.

Computational fluid dynamics (CFD) is a relatively new discipline that has existed for around the same amount of time as computing. This is where Beyond Math is making its initial progress. The mathematical formulas that determine an object's motion in air, water, or the air surrounding it are incredibly intricate. Although we've made steady progress in predicting, for example, the flow of air over a wing, we're still far from ideal, and the computational capacity required to achieve our current level of prediction is restricted to supercomputers and GPU clusters. As a result, there is a lot of waiting involved in the design process for products like automobiles, aircraft, and boats.

The objective of Beyond Math is to accelerate the digital design phase, i.e., reduce the time it takes from ideation to viability assessment. "They're asking me if changing this design will increase my car's fuel efficiency. Consider that you have six months to build a component for a plane. Considering how long a simulation takes, you might get 20 chances to try things out. But in the same six months, you might be able to run a million modifications if a designer has an idea and produces results in a matter of seconds or minutes, according to Garvey.

Several teams in Formula 1 racing are investigating the possibility of utilizing Beyond Math to speed up their aerodynamics and vehicle design procedures. This is one of the software's initial markets. They are among the biggest consumers of CFD, and because they move quickly, they will take up new technology. We have been actively collaborating with a few Formula One teams, conducting extensive assessments and comprehending their fundamental issues. A platform that will truly accelerate their autos is almost here, according to Garvey. In fact, he said that he hoped that within six months "we'll be able to show that customers are benefiting from these models, and they've gone out of research and proofs of concepts into things that have real impact," although he cautioned that this was no guarantee.

That should be made possible by recent funding: Beyond Math just completed a $8.5 million seed round headed by UP.Partners and including participation from Insight Partners and InMotion Ventures. With the purchase of Nvidia DGX 200s and collaboration with the chip giant on an intriguing new use case for its widely available compute technology, the startup plans to double the size of its workforce and increase its computational capacity.