• sales

    +86-0755-88291180

Hailo-8-Acce-A User Guide

Introduction

Hailo-8/8L chip with 26/13Tops computing power.

Feature

  • Neural network inference accelerator with 13/26 trillion operations per second (TOPS) built on Hailo-8/8L chips.
  • PCIE M.2 HAT+ for connecting the AI module to the Raspberry Pi 5.
  • Support installing with hardware kit.
  • Stackable GPIO pin header.

Hardware Connection

Pay attention to the cable orientation, as shown below:

Working with Raspberry Pi

Update

#1: update the software
sudo apt update && sudo apt full-upgrade
sudo rpi-eeprom-update

#Configure CLI (Not required for 2024 or later version systems)  
sudo raspi-config
#Advanced Options -> Bootloader Version, select Latest. Then, click Finish or press ESC key to exit raspi-config 

#2: Update firmware
sudo rpi-eeprom-update -a

Identify Device

1: Enable the PCIE interface:

Connect the hardware, and the latest system system will detect the hardware, connect the hardware will automatically enable the PCIE interface
If not, you can execute: add "dtparam=pciex1" in /boot/firmware/config.txt 

2: Enable PCIE Gen3, add the following content in /boot/firmware/config.txt (Gne3 mode must be enabled):

dtparam=pciex1_gen=3

3: Reset the PI5 after modification, and then the device will be identified. (You can not reboot first and then reboot when the library is installed.)


Test Demo

rpicam-apps runs camera demo with Hailo AI Neural Network Accelerator.

Preparation:

1: Raspberry Pi 5 and Hailo-8 Acce A kit
2: Install 64-bit Raspberry Pi OS Bookworm
3: Install Raspberry Pi camera (for testing, use Raspberry_Pi_Camera _Module_3 to connect the CAM1 interface.)


1: Install and use the required dependencies for AI Kit:

sudo apt install hailo-all

2: Reboot the device:

sudo reboot

3: Check whether the driver is normal:

hailortcli fw-control identify

Or execute "dmesg | grep -i hailo" to check the log 

4: Check the camera:

rpicam-hello -t 10s
Please ensure the camera is in normal operation 

5: Clone rpicam-apps:

git clone --depth 1 https://github.com/raspberrypi/rpicam-apps.git ~/rpicam-apps

6: Test:

Object test
rpicam-hello -t 0 --post-process-file ~/rpicam-apps/assets/hailo_yolov6_inference.json --lores-width 640 --lores-height 640
Yolov8 model
rpicam-hello -t 0 --post-process-file ~/rpicam-apps/assets/hailo_yolov8_inference.json --lores-width 640 --lores-height 640
YoloX model
rpicam-hello -t 0 --post-process-file ~/rpicam-apps/assets/hailo_yolox_inference.json --lores-width 640 --lores-height 640
Yolov5 human and facial models
rpicam-hello -t 0 --post-process-file ~/rpicam-apps/assets/hailo_yolov5_personface.json --lores-width 640 --lores-height 640

Image segmentation
rpicam-hello -t 0 --post-process-file ~/rpicam-apps/assets/hailo_yolov5_segmentation.json --lores-width 640 --lores-height 640 --framerate 20
Posture estimation
rpicam-hello -t 0 --post-process-file ~/rpicam-apps/assets/hailo_yolov8_pose.json --lores-width 640 --lores-height 640

For more details, you can refer to GitHubHailo official website.