Automating the Edge with Robotics

Ambient Scientific’s edge processor gears up for mass production

Categories Edge Computing News  |  Hardware
Ambient Scientific’s edge processor gears up for mass production

Ambient Scientific has announced volume production of its first AI System on Chip (SoC), GPX10. This chip is designed to run cloud-free AI applications on portable battery-powered devices.

Ambient Scientific is a fabless semiconductor company that develops processors for on-device and edge AI applications. The GPX10, manufactured using 40 nm technology, offers a software development kit with a neural network compiler. The evaluation kit will be available for purchase from Q1 2024.

Satish Kutty, the head of strategy and business development at Ambient Scientific, says: “We’re excited to be already partnering with customers to deliver AI applications powered by GPX-10, that includes a driver monitoring system, ECG wearables and always-on voice-based distress systems.”

The chip is compact and powered by DigAn technology for on-device and edge applications. It offers efficient AI throughput, privacy and security.

In combination with Ambient Scientific’s software toolchain, the GPX10 offers a programmable solution for applications in various industries. With its AI compiler, development libraries, and training toolkit, the company says the GPX10 enables sensor fusion, audio and vision applications.

Ambient Scientific’s founder and CEO, GP Singh, says his company has combined digital and analog computing concepts for scalable efficiency.

“Our processors are manufactured and can be programmed like any digital chip while carrying power and area efficiency of analog,” explains Singh.

GP Singh, the founder and CEO of Ambient Scientific, explains it is essential to unleash the potential of AI beyond the confines of the cloud and the internet. He states that to harness its power at the edge fully, a reimagination of conventional computing architecture is in order.

The team says they achieved this feat through innovation, including circuit technologies and a new Instruction Set Architecture (ISA). Features include 3D memory with in-memory computing for efficient operand disbursement, a multi-channel ultra-low-power sensor ADC for intelligent sensor functions and an always-on sensor fusion DMA for low-power AI applications.

Read more:

BrainChip, Prophesee to deliver “neuromorphic” event-based vision systems for OEMs

Arm offers smaller chip, better images for computer vision systems

Article Topics

 |   | 

Comments

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Featured Edge Computing Company

Edge Ecosystem Videos

Automating the Edge

“Automating

Deploying AI Models at the Edge

“Deploying

Latest News