Coral Dev Board Micro - development board with i.MX RT1176 microcontroller

Description

Coral Dev Board Micro is a board with i.MX RT1176 dual-core microcontroller, Coral Edge TPU, camera and microphone. i.MX RT1176 is a microcontroller developed by NXP Semiconductors. It is based on a dual-core architecture with ARM Cortex-M7 and ARM Cortex-M4, which run at a maximum frequency of 800 MHz (Cortex-M7) and 400 MHz (Cortex-M4). It has 2 MB of SRAM on board. The chip is designed for high-performance real-time applications. The board is also equipped with 64 MB of external SDRAM and 128 MB of NAND Flash memory.

The main application of the Coral Dev Board Micro is machine learning applications. The Coral Edge TPU is responsible for performing these functions. This is a small ASIC designed by Google that supports the fast performance of TensorFlow Lite models while maintaining energy efficiency. One Edge TPU is capable of performing 4 trillion operations per second (4 TOPS). The use of a dedicated ML gas pedal in the device reduces latency, increases data security and removes the need for a constant internet connection.

The module is designed for artificial intelligence applications using machine learning for image and speech processing. Therefore, the user has a QVGA (320x240 pixels) Himax HM01B0 sensor camera with a wide 110° viewing angle and a microphone with PDM output.

The module is powered by a USB Type-C connector, which also serves as data input/output with a computer. For rapid prototyping, the user can use four LEDs and a user button. The GPIO outputs also provide other interfaces used in microprocessor applications, including UART, I2C, SPI bus, as well as PWM outputs and analog inputs. In addition, two B2B (Board to Board) connectors are provided on the bottom of the board to allow the connection of additional expansion boards with Ethernet, SDIO, CSI, DSI, PDM, SAI, among others, as well as JTAG debug interface pins.

NXP's A71CH chip is responsible for data security, providing authentication of cryptographic devices, as well as secure sharing and storage of authentication data. With this board, you can build low-power systems with high-speed artificial neural network processing for machine vision and audio ML applications.

Features

• NXP i.MX RT1176 MCU:
  • ARM Cortex-M7:
    • 800 MHz
    • 32 KB L1 instruction cache
    • 32 KB of L1 data cache
    • Floating point unit (FPU) with single and double precision support with Armv7-M FPv5 architecture support
    • TensorFlow Lite support
  • ARM Cortex-M4:
    • 400 MHz
    • 16 KB instruction cache
    • 16 KB data cache
    • Single precision FPU defined by Armv7-M FPv4-SP architecture
    • TensorFlow Lite support
  • Memory:
  • 2 MB SRAM: 512 KB TCM for Cortex-M7, 256 KB TCM for Cortex-M4, 1.25 MB OCRAM
  • 4 KB of safe, always-on RAM
  • 256 KB boot ROM
• NXP A71CH: cryptographic device authentication
  • Coral Edge TPU coprocessor:
  • High inference performance for TensorFlow Lite models
  • 4 trillion operations per second (TOPS)
  • USB 2.0 interface with RT1176 MCU
• Camera:
  • Himax HM01B0 CMOS sensor
  • 324 x 324 px or QVGA (320 x 240) native resolution
  • FOV 110º
  • Focal ratio ƒ/2.0
  • Fixed focus
• PDM microphone
• External memory: 64 MB SDRAM, 128 MB Flash NAND
• GPIO connector 2 x 12 pins: GPIO, PWM, UART, I2C, SPI, DAC, ADC, GND, 3V3, 1V8, VSYS
• B2B connector 2 x 100 pins: support for Coral Wireless Add-on and Coral PoE Add-on
• USB Type C connector: power and USB communication
• 4 LEDS:
  • Green LED to indicate camera activity
  • Green LED programmable by user
  • Orange LED to indicate card operation status
  • White LED to indicate the operating status of the Edge TPU
• Buttons: reset and user button
• Dimensions: 65 x 30 mm