Automotive SoC combines 500 eTOPS with high-res image and radar processing

January 05, 2022 // By Christoph Hammerschmidt
Automotive SoC combines 500 eTOPS with high-res image and radar processing
The AI vision technology company Ambarella has introduced a scalable SoC family, the CV3, for use in autonomous vehicles. With a processing performance of up to 500 eTOPS, the manufacturer claims the pole position in the market for such devices. With its combination of multi-sensor perception, data fusion and path planning, the chip supports autonomous driving up to level 4.

The CV3 domain controller family represents a 42x increase over Ambarella’s prior automotive family. Featuring up to 16 Arm Cortex-A78AE CPU cores, the CV3 provides up to a 30x boost in CPU performance over the prior generation in support of autonomous vehicle (AV) software applications. According to Ambarella, this product family enables centralized, single-chip processing for multi-sensor perception, including high-resolution vision, radar and ultrasonic. In addition, it empowers deep fusion for multiple sensor modalities and AV path planning. The result is robust ADAS and L2+ to Level 4 autonomous driving (AD) systems with greater levels of environmental perception in challenging lighting, weather and driving conditions for both driver viewing and machine perception.

“With our CV3 AI domain controller family, we are now capable of running the full ADAS and AD stack with a single chip, while providing unprecedented performance and power efficiency,” says Ambarella CEO and President, Fermi Wang. “Through enhancements to our on-chip ISP, along with simultaneous radar processing that can take advantage of our Oculii adaptive AI algorithms, Ambarella is helping the automotive industry unlock greater levels of perception accuracy across all environmental conditions to realize the promise of autonomous driving.”

The CV3 marks the debut of Ambarella’s next-generation CVflow architecture, which continues the company’s algorithm-first design philosophy. This drove the development of the on-chip neural vector processor (NVP), with up to 500 eTOPS of AI compute, high power efficiency and support for the latest advancements in neural network (NN) inferences. The NVP is also enhanced to run advanced radar perception software, such as the Oculii adaptive AI software algorithms. This is complemented by a new floating-point general vector processor (GVP) designed to offload classical computer vision and radar processing from the NVP engines, and floating-point-intensive algorithms from the Arm CPUs.

The CV3 family’s hardware scalability allows customets to unify their software stacks across their entire retail portfolios, while reducing the cost and complexity of software development, Ambarella


Vous êtes certain ?

Si vous désactivez les cookies, vous ne pouvez plus naviguer sur le site.

Vous allez être rediriger vers Google.