Machine Learning On Raspberry Pi Pico with Edge Impulse

This repository runs an exported impulse on the Raspberry Pi Pico / RP2040. See the documentation at Running your impulse locally. This repository is based off of the Arducam Pico4ML Magic Wand Example.

Requirements

Hardware

• Raspberry Pi Pico.

Software

• Edge Impulse CLI.
• GNU ARM Embedded Toolchain.
• CMake.
• Rasperry Pi Pico SDK:

  git clone -b master https://github.com/raspberrypi/pico-sdk.git
  cd pico-sdk
  git submodule update --init
  export PATH="<Path to Pico SDK>:$PATH"

Building the application

Get the Edge Impulse SDK

Unzip the deployed C++ library from your Edge Impulse project and copy to the source directory of this repository:

example-standalone-inferencing-pico/
├─ source
├─- model-parameters
├─- edge-impulse-sdk
├─- tflite-model
├─- CMakeLists.txt
├─ .gitignore
├─ LICENSE
├─ README.md
└─ pico_sdk_import.cmake

Compile

1. Create the build folder:

   mkdir build && cd build

2. Compile:

   cmake ..
   clear && make -j4

Flash

Connect the Raspberry Pi Pico to your computer using a micro-USB cable while pressing and holding the BOOTSEL button.

Drag and drop the build/pico_standalone.uf2 file to the RPI-RP2 disk in your file explorer.

Serial connection

Use screen, minicom or Serial monitor in Arduino IDE to set up a serial connection over USB. The following UART settings are used: 115200 baud, 8N1.