Toyota Research Institute Empowers Vehicle Designers with AI Tools

Avinash Balachandran, director of TRI’s Human Interactive Driving Division, highlighted the limitations of generative AI tools in handling complex engineering and safety considerations in car design. 

“This technique combines Toyota’s traditional engineering strengths with the state-of-the-art capabilities of modern generative AI,”

said Balachandran, whose team contributed to the development of the technology.

TRI researchers have developed an approach that combines engineering constraints with the design process. This technique considers factors such as “drag (which affects fuel efficiency) and chassis dimensions like ride height and cabin dimensions (which affect handling, ergonomics, and safety).”

The team merged optimization theory and text-to-image-based generative AI to create an algorithm that enables designers to optimize engineering constraints using text-based stylistic prompts.

As an example, a designer can request a set of designs based on an initial prototype sketch, specifying stylistic properties like “sleek,” “SUV-like,” and “modern” while optimizing a performance metric. The research paper focuses on aerodynamic drag, but the approach can also optimize other performance metrics or design constraints inferred from an image.

Last year, Toyota announced its entrance into the metaverse. The Japanese carmaker conducted employee meetings and created virtual workspaces in the digital world. The company believes that virtual workspaces will provide a more immersive mode of communication where employees can engage with each other using avatars.

Read more: