Google edge AI board taps NXP’s i.MX micro
Google is launching an edge AI board in the same form factor as a Raspberry Pi Nano for its Coral AI processor and NCP’s i.MX microcontroller for the first time.
The Coral Dev Board Micro is a microcontroller board with a built-in camera, microphone, and Coral Edge TPU, allowing developers to quickly prototype and deploy low-power embedded systems with on-device ML inferencing.
Development is focussed on the NXP i.MX RT1176 microcontroller with ARM Cortex-M7 and Cortex-M4 cores for low-power machine learning inferencing of TensorFlow and TensorFLowLite using the Coral accelerator chip that provides 4 TOPS of 8bit performance at 2W.
- Free custom AI boards for Google Coral chip
- AI accelerator boosts Matter-ready IoT chip
- €100m project to develop low power edge AI microcontroller
The board has 512 Mbit of SRAM and 1Gbit of flash with a 324 x 324 pixel colour camera and PDM mono microphone. There are two 12-pin GPIO headers and two 100-pin board-to-board connectors as well as a USB Type-C (USB 2.0) interface.
Accessories include a Wi-Fi/Bluetooth Module Board and an Ethernet/Power-Over-Ethernet Board
- Google machine learning TPU available on Coral board
- World’s smallest AI module with Coral intelligence
- Murata and Google team to develop ultrasmall AI module
- Fraunhofer extends RISC-V embedded processor for edge AI
- Finnish consortium tapes out multicore RISC-V edge AI IoT chip
Other articles on eeNews Europe