Deploy ML models to Arm edge devices using Edge Impulse and AWS IoT Greengrass
Introduction
Understand the Edge Impulse and AWS IoT Greengrass deployment architecture
Select and set up your edge device
Set up your Edge Impulse project
Install AWS IoT Greengrass on your Arm edge device
Store your Edge Impulse API key in AWS Secrets Manager
Create the Edge Impulse Greengrass component
Deploy the component to your edge device
Verify inference and view results
Next Steps
Deploy ML models to Arm edge devices using Edge Impulse and AWS IoT Greengrass
Introduction
Understand the Edge Impulse and AWS IoT Greengrass deployment architecture
Select and set up your edge device
Set up your Edge Impulse project
Install AWS IoT Greengrass on your Arm edge device
Store your Edge Impulse API key in AWS Secrets Manager
Create the Edge Impulse Greengrass component
Deploy the component to your edge device
Verify inference and view results
Next Steps
Choose your platform
Before you can install AWS IoT Greengrass and deploy an Edge Impulse model, you need a Linux-based Arm device to act as your edge device. You can use one of four platform options. Select the one that matches your available hardware and follow the setup instructions.
If you don’t have any of the supported physical hardware boards, use the Amazon EC2 option. By following the Amazon EC2 instance setup instructions, you can create an Arm-based virtual machine in the cloud that behaves like a local edge device. With an EC2 instance, you can complete every step in the Learning Path without dedicated hardware.
When you set up your platform, you’ll install the required dependencies such as build tools, Node.js, GStreamer, and Java. The instructions also include a device-specific JSON configuration that you’ll use later when deploying the Greengrass component.
Arm-based Amazon EC2 instance (no hardware required)
Use an Arm-based Amazon EC2 instance option if you don’t have a physical edge device. Create an Ubuntu-based EC2 instance with an Arm processor (Graviton) that simulates a local edge device. Because there’s no camera attached, you’ll use a sample video file for inference input.
Raspberry Pi 5 with Raspberry Pi OS
The Raspberry Pi 5 is a widely available, affordable Arm board with full Edge Impulse and Greengrass support. You can run inference with an attached USB camera or use a sample video file.
NVIDIA Jetson with JetPack 5.x or 6.0
If you have an NVIDIA Jetson board such as Nano, Xavier, or Orin, you can take advantage of GPU-accelerated inference. This option assumes JetPack is already flashed onto the device.
Qualcomm QC6490 with Ubuntu
For Qualcomm QC6490-based development boards running Ubuntu, you can use both the on-board Qualcomm camera and USB-attached cameras. You can also use file-based inference without a camera.
What you’ve accomplished and what’s next
You’ve selected an Arm edge platform, completed its setup, and saved the device-specific component configuration.
Next, create your Edge Impulse project and build a model deployment.