Synopsys, Sima.ai team to build automotive edge AI silicon
Synopsys and Sima.ai are developing an end-to-end workflow for workload-specific silicon and software for next generation automotive edge AI.
The two are combining the Synopsys EDA tools, automotive-grade IP, and hardware-assisted verification with SiMa.ai’s machine learning accelerator (MLA) IP and complete ML software stack application development environment for maximum customization of IP, subsystems, chiplets, and SoCs.
Advanced Driver Assistance Systems (ADAS) and In-vehicle Infotainment (IVI) are key differentiators for car makers and the advent of generative AI is driving the need for local AI processing. However, current software-defined vehicle (SDV) architectures are not equipped to support related demands for the diversity of applications and the required compute, performance, and reliability.
AI-ready, workload-verified, power-efficient software architectures need hardware-software co-design solutions, from silicon –to–systems, to reduce development costs and de-risk the start-of-production timelines.
- Synopsys buys Imperas for verification tools and models
- Rapidus signs 2nm deals with Synopsys, Cadence
- TUV certifies Synopsys ARC core for automotive cybersecurity
Synopsys has over 50 OEMs and Tier 1 suppliers using its virtual prototyping technologies, and more than 1,000 virtual models have been developed from SoC IP to board-level ICs and ASICs.
The SiMa.ai tech stack emphasizes flexibility, offering support for a wide range of models, sensors and applications, and automatically optimizing them for peak performance.
The integrated Synopsys and SiMa.ai workflow will be boost early architecture exploration with the ability to to guide automotive design engineers to optimally-choose performance, power, and software application requirements for custom or 3rd party SoC development.
The workflow will be optimized for multi-modal functions and GenAI and include virtual prototyping and emulation to accelerate software/hardware integration using silicon-proven building blocks and subsystems for high-performance automotive edge AI SoCs and chiplets.
“Our collaboration with SiMa.ai will provide automotive companies a significant step forward in modernizing their hardware/software co-design processes to meet demand for advanced, in-car experiences that are increasingly costly and complex to deliver,” said Ravi Subramanian, head of the Synopsys Product Management and Markets Group. “Our leadership in architecture exploration, IP, and hardware-assisted verification, combined with SiMa.ai’s innovative performance and power-optimized ML capabilities, will enable customers to differentiate while continuing to meet stringent cost considerations and industry standards.”
“Collaborating closely with Synopsys allows us to provide automotive manufacturers with complete, optimized solutions that accelerate their development cycles,” said Krishna Rangasayee, Founder and CEO, SiMa.ai.
“The SiMa.ai MLSoC platform was designed for best possible performance at lowest power consumption across a diverse set of applications—that was the missing piece in the puzzle to make future cars so much smarter and safer than what we have today. When combined with Synopsys’ industry-leading IP and software, we will deliver a powerful foundation for innovation across auto OEMs in autonomous driving and in-vehicle experiences.”
If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :
