Google Coral Dev Board AI 開發板 Google AI 晶片Edge TPU 支援 TensorFlow

描述

Google Coral Dev Board AI 開發板 Google AI 晶片Edge TPU 支援 TensorFlow

Google Coral Dev Board AI 開發板 採用 NXP i.MX 8M SOC，具有4個Cortex-A53處理器核心，與GC7000繪圖處理器，1GB LPDDR4記憶體與8GB eMMC儲存媒體，可以透過microSD擴充容量，並搭載Edge TPU為深度學習人工智慧加速。此外板上還有USB 3.0 Type-A、USB Type-C、USB Type-C電源輸入、HDMI 2.0、GbE乙太網路等端子，並具有Wi-Fi5（IEEE 802.11b/g/n/ac，支援2.4/5GHz雙頻，2×2 MIMO）與藍牙4.1無線通訊。其作業系統目前僅支援Debian，能夠執行TensorFlow Lite框架，未來將支援Python、C++等程式語言。

Coral平台不但能協助開發人員快速打造原型，還標榜很容易就可擴大至生產線，因為Google也提供了模組系統及PCIe版本的加速器供大量採購。 Coral Dev Board 就像一塊專為本地 AI 優化的 Raspberry Pi ，附有一片可置換、備有 Edge TPU 機械學習加速器的 SoM 系統模組，它使用名為 Mendel Linux 的 OS ，支援 TensorFlow Lite 框架。編程語言方面，現在只支援 Python ，但將來就會加入 C++ 支援。

Google Coral Dev Board AI 開發板 支援 TensorFlow 與 TensorFlow Lite

軟體工具則主要源自 TensorFlow 與 TensorFlow Lite，其中的TensorFlow Lite模型必須先行量化再與Google的工具鏈編譯，才能直接於Edge TPU運作。且Google提供了十多種已訓練及編譯完畢的模型供開發人員參考。

Google Coral Dev Board AI 開發板硬體規格參數

概述

Google Coral Dev Board AI 開發板 是一款單板計算機，當您需要以小巧的外形進行快速機器學習（ML）推理時，它是理想選擇。您可以使用開發闆對嵌入式系統進行原型設計，然後使用板載 Coral System-on-Module（SoM）與您的定制PCB硬件相結合，進行生產。

SoM提供完全集成的系統，包括恩智浦的iMX8M片上系統（SoC），eMMC內存，LPDDR4內存，Wi-Fi和藍牙，但其獨特的功能來自谷歌的Edge TPU協處理器。Edge TPU是由Google設計的小型ASIC，可以低功耗提供高性能ML推理。例如，它可以以功率有效的方式執行最先進的移動視覺模型，例如400 FPS的MobileNet v2。

基板提供了項目原型所需的所有外圍設備連接，包括USB 2.0 / 3.0端口，DSI顯示接口，CSI-2攝像頭接口，以太網端口，揚聲器端子和40針I / O接頭。

技術文件與資源

• Google 原廠教學文件
• 相關範例專案與教育

Google Coral Dev Board

The Coral Dev Board is a single-board computer with a removable system-on-module (SOM) that contains eMMC, SOC, wireless radios, and Google’s Edge TPU. It’s perfect for IoT devices and other embedded systems that demand fast on-device ML inferencing.

 

You can use the Dev Board as a single-board computer for accelerated ML processing in a small form factor, or as an evaluation kit for the SOM that’s on-board. The 40 mm × 48 mm SOM on the Dev Board is available at volume. It can be combined with your custom PCB hardware using board-to-board connectors for integration into products.

 

The SOM is based on NXP’s iMX8M system-on-chip (SOC), but its unique power comes from the Edge TPU coprocessor. The Edge TPU is a small ASIC designed by Google that provides high performance ML inferencing with a low power cost. For example, it can execute state-of-the-art mobile vision models such as MobileNet v2 at 100+ fps, in a power efficient manner.

Edge TPU key benefits

• High-speed TensorFlow Lite inferencing
• Low Power
• Small footprint

The baseboard includes all the peripheral connections you need to prototype a project, including USB 2.0/3.0 ports, DSI display interface, CSI2 camera interface, Ethernet port, speaker terminals, and a 40-pin GPIO header.

Coral is a division of Google, that helps you build intelligent ideas with our platform for local AI.

Features

SOM

• NXP i.MX 8M SOC (Quad-core Cortex-A53, plus Cortex-M4F)
• Google Edge TPU ML accelerator coprocessor
• Cryptographic coprocessor
• Wi-Fi 2×2 MIMO (802.11b/g/n/ac 2.4/5GHz)
• Bluetooth 4.1
• 8GB eMMC
• 1GB LPDDR4

USB Connections

• USB Type-C power port (5V DC)
• USB 3.0 Type-C OTG port
• USB 3.0 Type-A host port
• USB 2.0 Micro-B serial console port

Audio connections

• 3.5mm audio jack (CTIA compliant)
• Digital PDM microphone (x2)
• 2.54mm 4-pin terminal for stereo speakers

Video connections

• HDMI 2.0a (full size)
• 39-pin FFC connector for MIPI-DSI display (4-lane)
• 4-pin FFC connector for MIPI-CSI2 camera (4-lane)

MicroSD card slots

Gigabit Ethernet ports

40-pin GPIO expansion headers

Supports Debian Linuxs

Models are built using TensorFlow

Fully supports MobileNet and Inception architectures though custom architectures are possible

Compatible with Google Cloud

Supports TFLite

No need to build models from the ground up. TensorFlow Lite models can be compiled to run on the Coral Dev Board.

Scale from prototype to production

Considers your manufacturing needs. The SOM can be removed from the baseboard, ordered in bulk, and integrated into your hardware.

Includes a full system

SOC + ML + Connectivity all on the board running a variant of Linux, so you can run your favourite Linux tools with this board.