Can you tell us a bit about Monolith and how its technology is helping automakers?
Monolith enables engineers to use AI to solve their most intractable physics problems. We do this by enabling engineers to use their existing data from the current product development process.
Using their existing data, they can build machine learning models to immediately predict outcomes for challenges that would otherwise have to be solved using extensive, time-consuming physical testing. These machine learning models learn from existing data and identify patterns to solve challenges that have hitherto been beyond the scope of what the human engineer or simulation technology can do. Instead of running expensive, time-consuming tests to eliminate this uncertainty, engineers can work with the Monolith platform to quickly understand and predict performance without having to program or code the models.
And because Monolith is purpose-built for domain experts rather than statisticians or expert coders, our no-code interface allows engineers to use their expertise to solve highly complex problems that take advantage of the wealth of data that already exists. and removes dependence on other teams less familiar with the data or problem.
What previously impossible R&D challenges can AI help solve now?
As an engineer, you may have been tasked with building a model and then using that model to virtually design the perfect product in the hopes that the physics-based simulation approach will do its job. But what I’ve found is that if you want to radically accelerate the speed of developing a new product, you need a radically different solution to understand the physical challenges that are still not fully understood.
One of our customers in the industrial space is doing just this. Engineers at Honeywell previously used Computational Fluid Dynamics (CFD) simulations to understand the complexities of gas fluid dynamics when developing their smart gas meter, but the simulations were not 100% accurate, creating a critical gap in understanding between simulation and reality . Using Monolith, the engineers use statistical machine learning methods to close the gap, allowing them to immediately and accurately understand the impact of different temperature conditions and types of gases in all operating conditions, including extreme and unstable parameters. As a result, Honeywell’s engineering team builds superior products on significantly shorter timeframes.
The ideal use case for AI is when engineers are trying to understand the physics of complex systems that cannot be fully represented by simulation and therefore require significant physical testing to calibrate. When faced with intractable physics problems, they can use Monolith to immediately leverage their existing data and immediately solve the previously unsolvable problems, literally giving them weeks or months of their time.
Can you give us some specific examples of how OEMs are using AI in the R&D process?
By training the machine learning Monolith models with the company’s test data, BMW engineers are using AI to solve previously intractable physics challenges and predict the performance of highly complex systems.
We started working with BMW Group’s crash test engineering team in 2019 to see if AI could predict crash performance and, most importantly, do it significantly earlier in the vehicle development process. BMW engineers built machine learning models using the wealth of their existing crash data and were able to accurately predict the force on the human tibia for a range of different types of crashes without doing any physical crashes.
Going forward, the accuracy of the machine learning models will continue to improve as more data becomes available and the platform is further embedded into BMW’s engineering workflow. This means engineers can optimize crash performance earlier in the design process and reduce reliance on time-consuming, costly testing, while making historical data infinitely more valuable.
Should AI be seen as a threat to engineers and possibly their jobs?
Quite the opposite. One of our clients recently referred to Monolith as Augmented Intelligence as it increases the technical expertise of his team. It is an ideal tool for engineers who are short on time and who are frankly under enormous pressure to develop the next generation of vehicles.
With Monolith, an automotive customer reported a 70% reduction in track test time, plus a reduction in overall costs of up to 50%. This is just one example of how the efficiencies achieved by AI free up time that can be spent refining even better products. As one Honda executive said of our technology, “It almost gives us superpowers.”
Do you see emerging trends where AI can help?
Whatever new technology is introduced, whether autonomous, connected or electric, engineers will always have to create a fundamentally great car to stay competitive and drive demand – from first-class acoustics and better fuel economy to safety and vehicle dynamics.
AI technology can revolutionize vehicle development by enabling engineers to gain the best possible insights and predict outcomes from existing engineering data, much earlier in the development process. This allows engineers to make design and engineering decisions faster and more efficiently, giving them time to explore even more design parameters and operating conditions.
Ultimately, this means OEMs can bring better vehicles to market faster, which is vital not only to achieving our shared EV ambitions, but also enabling engineers to do what they love to do: develop incredible products.
We hear that while manufacturers remain enthusiastic about automated driving, the challenge will take more time and effort to fully realize. What is your view on the road to autonomous driving?
Correct. Last year, AV companies Waymo and Cruise cumulatively drove 3.2 million miles to vigorously train and test their autonomous technology. Engineering teams are barely getting any more money this time around – competitors are moving too fast – and yet the complexity of vehicle development has never been greater.
When engineers struggle to understand the tenacious physics of complex vehicle systems, machine learning models can complement the solving of the underlying physics with AI-based statistical predictions. Autonomous technology is the perfect example of where a lot of data has been created to understand a complex system.
OEMs can use AI to extract existing engineering data from simulations or physical tests to build machine learning models that immediately predict further test results or uncover new insights buried in the data.
Do you expect to see a greater role for simulation and AI as part of training, validating and testing AVs?
Yes, but we understand that introducing new tools into an age-old product development process requires vision, courage and experimentation. That experimentation becomes much easier when engineers understand that AI is not a complete replacement for simulations and physical testing – rather, it is the essential piece of the puzzle to enable engineers to achieve design convergence much faster and more efficiently.
The introduction of AI into modern vehicle development is a similar incremental change: engineers suddenly no longer need to solve the physics underlying a mathematical system, but can now immediately access the insights already hidden in their existing engineering data. .
What does the future hold for Monolith?
The world’s best engineering teams, from Rolls Royce to Honda and Siemens, use Monolith to reduce product development and create even higher quality products. Specifically in the automotive sector, engineering teams are bringing Monolith into more and more technical functions that generate massive amounts of data, from crash testing to aerodynamics, motorsports and, as mentioned, ADAS.
We are ready to scale quickly, powered by the globally renowned engineering teams from major OEMs who adopt our technology and extensive IP portfolio.