In the same way many carmakers have design studios in Southern California, they also have tech labs in Silicon Valley. There, engineers play with latest developments in the digital world. Nissan recently revealed that their offices in the Bay Area have been using artificial intelligence to fast-track research in automotive materials.
AI is a popular buzzword right now, but it can do more than write term papers for students and pretend to be celebrities. According to Nissan, the researching and testing of new materials for car parts used to take 20 years, with numerous cycles of trial and error. Now, AI simulations have cut that time down to just two years — 1,000 times faster.
“With machine learning and AI, you can simulate the properties of materials for a lot more cases than you can test experimentally, in a short time,” said Bala Radhakrishnan, principal researcher, simulation. “It can help you sample millions of materials, and then screen for candidates based on the properties that you want.”
Nissan explains that the AI they use is different from something like ChatGPT. ChatGPT is a form of generative AI, which creates new material by scraping existing knowledge off the web. Nissan’s AI is a form of machine learning, which simulates millions of, say, automotive materials, and runs tests on them to predict the most ideal properties — strength, temperature sensitivity, conductivity — for their applications.
Humans are still an important part of the process. Once AI filters the best candidates, engineers still interpret the data and test the materials in the physical world. AI is merely a form of helping predict the most likely candidates and strip out the noise.
In practical terms, one of the technologies AI has helped accelerate is solid-state batteries for electric cars. Nissan plans to bring those to market by 2028.
