Arm accelerates Edge AI with Ethos-U NPU and IoT reference design

NPU said to deliver 4x performance uplift for high performance edge AI applications such as factory automation and smart cameras through new IoT reference design platform to accelerate deployment of voice and vision systems

Noting that as edge artificial intelligence (AI) scales, silicon innovators must navigate growing system and software complexity, while software developers need more consistent, streamlined experiences and easy integration with emerging AI frameworks and libraries, leading chip design firm Arm has unveiled its third-generation NPU to support edge AI, the Arm Ethos-U85.

In addition, a new Arm internet of things (IoT) reference design platform, Corstone-320, brings together embedded IP with virtual hardware to accelerate deployment of voice, audio and vision systems.

Arm believes that the IoT and embedded industry is looking to the company to scale the edge AI opportunity. The new platform is said to deliver a 4x performance uplift and 20% higher power efficiency compared with its predecessor while scaling from 128 to 2048 MAC units (4 TOPs @1GHz). The Ethos-U85 is designed to address applications where Arm sees greater performance demands such as factory automation and commercial or smart home cameras.

Arm said that the new Ethos-U85 offers the same consistent toolchain so partners can take advantage of  existing investments to deliver a “seamless” developer experience. It provides support for AI frameworks such as TensorFlow Lite and PyTorch and supports Transformer Networks as well as convolutional neural networks (CNNs) for AI inference.

The company believes that transformer networks will drive new applications, particularly in vision and generative AI use cases for tasks like understanding videos, filling in missing parts of images or analysing data from multiple cameras for image classification and object detection.

After recognising the deployment of microprocessors into more high-performance IoT systems for use cases such as industrial machine vision, wearables and consumer robotics, Arm said that it has designed the Ethos-U85 to work with its Armv9 Cortex-A CPUs, to accelerate machine learning (ML) tasks and bring power-efficient edge inference into a broader range of higher-performing devices.

“Machine learning workloads for the next generation of edge AI applications demand high performance in a power efficient manner,” said Reza Kazerounian, co-founder and president, Alif Semiconductor. “Alif was the first to market with an edge AI solution based on Arm Cortex-M55 and Ethos-U55, and we welcome Arm’s latest AI technology, Ethos-U85, which will deliver the compute performance required for our next generation Ensemble family of microcontrollers and fusion processors to address future edge AI and vision use cases.”

The Ethos family of NPUs has been licensed by more than 20 partners to date, and early adopters of the new Ethos-U85 include Alif and Infineon.

Edge AI use cases are becoming increasingly sophisticated and require secure, high performance compute systems to deliver on the opportunities of the AI era,” added Steve Tateosian, senior vice-president of industrial MCUs, IoT, wireless and compute business at Infineon. “We look forward to building on Infineon’s long-standing partnership with Arm and enabling these sophisticated systems with Arm Ethos-U85 and the transformer network support it provides for edge AI deployments.”

The Corstone-320 IoT reference design platform is said to be able to deliver the performance required to span the broad range of edge AI applications for voice, audio and vision, such as real time image classification and object recognition, or enabling voice assistants with natural language translation on smart speakers. The platform includes software, tools and support including Arm virtual hardware.

The company is confident that such a combination of hardware and software will accelerate product timelines by enabling software development to start ahead of silicon being available, rapidly improving time to market for increasingly complex edge AI devices.

Read more about Edge AI

Read more on Internet of Things (IoT)

Data Center
Data Management