---

ROS2 Driver for TI IWR6843 Mobile Tracker

A lightweight, Python-based ROS2 driver designed specifically for the Texas Instruments IWR6843 mmWave Radar running the Mobile Tracker firmware .

🚀 Motivation

While Texas Instruments provides an official ROS driver (ti_mmwave_rospkg), it is primarily designed for the "Out-of-Box" demo firmware. It does not correctly parse the specific output data formats (TLVs) generated by the Mobile Tracker firmware.

This project bridges that gap by providing a driver capable of:

1. Configuring the radar with a specific .cfg file (e.g., Mobile_Tracker_car.cfg).
2. Parsing the specific TLVs (Point Cloud, Side Info, Target List, Target Index) used in tracking applications.
3. Publishing standard ROS2 messages (PointCloud2) and TI-specific tracking messages.
4. Visualizing tracking results with MarkerArray in RViz.

📦 Dependencies

System Dependencies

• ROS2 (Tested on Humble with Ubuntu 22.04)
• Python 3
• PySerial: pip3 install pyserial
• NumPy: pip3 install numpy

ROS Dependencies

This driver relies on message definitions usually found in the TI ROS ecosystem. Ensure you have the message package compiled in your workspace:

• ti_mmwave_rospkg_msgs (Contains RadarTrackArray and RadarTrackContents)
• sensor_msgs
• visualization_msgs

⚙️ Configuration: Serial Ports

The IWR6843 typically exposes two serial ports:

1. Command/CLI Port (User UART): Used to send configuration commands.
2. Data Port (Data UART): Used to receive high-speed radar data.

You have two options to configure these ports:

Option 1: Using Udev Rules (Recommended)

To ensure the ports always map to consistent names (e.g., /dev/ttyUSB_RADAR_CLI and /dev/ttyUSB_RADAR_DATA) regardless of plug-in order, you should create a udev rule.

1. Check your device ID (usually 10c4:ea70 for Silicon Labs CP210x):

   Bash

   lsusb
   

2. Create a rule file:

   Bash

   sudo nano /etc/udev/rules.d/99-mmwave-radar.rules
   

3. Add the following lines (adjust idVendor and idProduct if yours differs):

   Plaintext

   # TI mmWave Radar CLI Port (usually the first interface, e.g., if00)
   SUBSYSTEM=="tty", ATTRS{idVendor}=="10c4", ATTRS{idProduct}=="ea70", ATTRS{interface}=="Silicon Labs Dual CP2105 USB to UART Bridge: Interface 0", SYMLINK+="ttyUSB_RADAR_CLI"
   
   # TI mmWave Radar DATA Port (usually the second interface, e.g., if01)
   SUBSYSTEM=="tty", ATTRS{idVendor}=="10c4", ATTRS{idProduct}=="ea70", ATTRS{interface}=="Silicon Labs Dual CP2105 USB to UART Bridge: Interface 1", SYMLINK+="ttyUSB_RADAR_DATA"
   

   Note: The interface names might vary. You can use udevadm info -a -n /dev/ttyUSBx to find unique attributes.

4. Reload rules:

   Bash

   sudo udevadm control --reload-rules && sudo udevadm trigger
   

Option 2: Specifying Ports via Parameters

If you do not wish to use udev rules, you can simply identify which port is which (e.g., /dev/ttyUSB0 and /dev/ttyUSB1) and pass them as parameters when running the node.

🛠️ Installation

1. Clone this repository into your ROS2 workspace src folder:

   Bash

   cd ~/ros2_ws/src
   git clone https://github.com/YOUR_USERNAME/mobile_tracker.git
   

2. Build the workspace:

   Bash

   cd ~/ros2_ws
   colcon build --packages-select mobile_tracker
   source install/setup.bash
   

▶️ Usage

Run with Default Settings

By default, the driver looks for /dev/ttyUSB0 (CLI) and /dev/ttyUSB1 (Data).

Bash

ros2 run mobile_tracker mmwave_6843_driver

Run with Custom Ports (Command Line)

If your ports are different (or if you mapped them via udev):

Bash

ros2 run mobile_tracker mmwave_6843_driver --ros-args -p cli_port:=/dev/ttyUSB_RADAR_CLI -p data_port:=/dev/ttyUSB_RADAR_DATA

Run with a Custom Config File

The driver loads the configuration file located in cfg/Mobile_Tracker_car.cfg by default. You can specify a different path:

Bash

ros2 run mobile_tracker mmwave_6843_driver --ros-args -p config_file:=/path/to/your/profile.cfg

📡 Topics Published

Topic Name	Message Type	Description
/ti_mmwave/radar_scan_pcl	sensor_msgs/PointCloud2	The raw point cloud data (X, Y, Z, Doppler, SNR).
/ti_mmwave/radar_track_array	ti_mmwave_rospkg_msgs/RadarTrackArray	List of tracked objects (ID, Position, Velocity, Acceleration).
/ti_mmwave/radar_track_marker	visualization_msgs/MarkerArray	Visualization markers for RViz (Cubes for objects, Arrows for velocity, Text for IDs).

📊 Visualization (RViz)

1. Open RViz2: ros2 run rviz2 rviz2
2. Set the Fixed Frame to ti_mmwave_0.
3. Add a PointCloud2 display and subscribe to /ti_mmwave/radar_scan_pcl.
4. Add a MarkerArray display and subscribe to /ti_mmwave/radar_track_marker.