AU 4D Radar

Introduction

au_4d_radar is a collection of ROS2 packages supporting 4D Radar Sensors. This program has been tested on ubuntu 22.04 ROS2 humble.
Supports automatic connection and communication functions without having to set the IP of each Radar sensor in a local network environment.

How to build the au_4d_radar

Create a new work directory of ROS2:

mkdir -p ~/share/ros2_ws/src

Download AU 4D Radar ROS2 Source code:

$ export VERSION=v2.2
$ cd ~/share/ros2_ws/src
$ git clone https://github.com/auradar/au_4d_radar.git
$ cd au_4d_radar
$ git checkout $VERSION
$ git submodule update --init --recursive

Download radar messages:

$ cd ~/share/ros2_ws/src
$ git clone https://github.com/ros-perception/radar_msgs.git

Download monitor messages:

$ cd ~/share/ros2_ws/src
$ git clone -b v2.0 https://github.com/auradar/mon_msgs.git

Install FlatBuffers:

1. Download the FlatBuffers

  export VERSION=v24.3.25
  git clone https://github.com/google/flatbuffers.git flatbuffers_$VERSION
  cd flatbuffers_$VERSION
  git checkout $VERSION
  git submodule update --init --recursive

2. Build and install the FlatBuffers

  mkdir build
  cd build

  # Configure to build both static and shared libraries
  cmake .. -DCMAKE_BUILD_TYPE=Release \
        -DFLATBUFFERS_STATIC_FLATC=ON \
        -DFLATBUFFERS_BUILD_SHAREDLIB=OFF \
        -DFLATBUFFERS_BUILD_CPP17=ON \
        -DCMAKE_INSTALL_PREFIX=/usr/local

  make -j$(nproc)
  sudo make install

Frame ID Naming convention:

The system_info.yaml file is structured as follows.

radars:
frame_id: identifier  

If you want to display an identifier instead of frame_id, enter frame_id: identifier in the system_info.yaml file.

radars:
27c06058: FRONT RIGHT
0b089dfa: REAR LEFT

Compile:

$ cd ~/share/ros2_ws
$ colcon build --cmake-args -DCMAKE_BUILD_TYPE=Relaese --packages-up-to au_4d_radar

Running Radar Node using launch actions:

$ cd ~/share/ros2_ws
source install/local_setup.bash
ros2 launch au_4d_radar run_radar.launch.py

Run-time composition using ROS services

Run in shell 1:

In the first shell, start the component container:

$ cd ~/share/ros2_ws
source install/local_setup.bash
ros2 run rclcpp_components component_container

Run in shell 2:

In the second shell load au_4d_radar node

$ cd ~/share/ros2_ws
source install/local_setup.bash
ros2 component load /ComponentManager au_4d_radar au_4d_radar::device_au_radar_node