Synopsys, SiMa.ai To Collaborate On AI Car Chips

Silicon Valley-based chip design automation firm Synopsys and AI start-up SiMa.ai are to work together to speed up the development of AI chips designed especially for automobiles, the companies said.

The deal is the latest to take advantage of the wave of interest in generative AI that has followed OpenAI’s public launch of ChatGPT in late 2022, and forsees tools such as voice assistants being integrated into cars in the next few years.

Synopsys is one of the industry’s leading electronic design automation (EDA) vendors, focusing on tools to help chipmaking companies design and verify new products.

SiMa.ai is focused on bringing a range of AI technologies to edge platforms such as cars, which require a combination of performance and energy efficiency.

Edge hardware

While initially focused on computer vision challenges, the start-up’s intellectual property is also geared toward integrating tools such as voice assistants into embedded hardware such as that which runs in cars.

The companies said they would work together on a new offering to accelerate the development of workload-specific hardware and software for AI features in next-generation automobiles.

The planned system is to include Synopsys’ EDA tools along with SiMa.ai’s intellectual property related to its machine learning acceleration and complete machine learning software stack application development environment.

Customers will be able to use the tools to develop custom offerings for subsystems, chiplets and systems-on-a-chip (SoCs), the firms said.

The companies said the platform would help customers develop AI-ready, workload-verified, power-efficient software architectures for vehicles.

Advanced driver-assistance features and in-vehicle information systems are becoming an important selling point for carmakers, especially with the rise of electric vehicles, which tend to emphasise high-tech systems.

Next-generation cars

To date, systems such as generative AI require fast internet connections linking to powerful data centres, an experience carmakers are working to adapt to operate in vehicles.

“Our leadership in architecture exploration, IP, and hardware-assisted verification, combined with SiMa.ai’s innovative performance and power-optimised ML capabilities, will enable customers to differentiate,” said Ravi Subramanian, head of Synopsys’ product management and markets group.

SiMa.ai’s MLSoC platform was designed for “best possible performance at lowest power consumption” to make cars “much smarter and safer than what we have today”, said SiMa.ai founder and chief executive Krishna Rangasayee.