Description

ROS2 node to process point cloud data from a radar system and calculate the corresponding body frame velocity. This process involves performing non-linear optimization using the point cloud data as inputs. To ensure accuracy, outliers are managed through the application of robust error models.

The node has been tested with the Smartmicro DRVEGRD 152 Front Radar and was designed for integration with the MicroStrain CV7-INS, but can applied for general use.

Building from Source

1. Install ROS2 and Create a Workspace
2. Clone the Repository into your workspace:

   git clone https://github.com/microstrain-robotics/radar_velocity_estimation.git --recursive
   

3. Install rosdeps for all the packages: rosdep install --from-paths ~/your_workspace/src -i -r -y
4. Build your workspace

   cd ~/your_workspace
   colcon build
   

Running the node

source /your_workspace/install/setup.bash
ros2 run radar_velocity_estimation_node radar_velocity_estimation_node

ROS Interfaces

Topic	Type	Description
/smart_radar/can_targets_0	sensor_msgs/PointCloud2	Input 4D radar point cloud with speed
/cv7_ins/ext/velocity_body	geometry_msgs/TwistWithCovarianceStamped	Output estimated bodyframe velocity with covariance
/radar_velocity_viz	geometry_msgs/TwistStamped	Output estimated bodyframe velocity, primarily used for visualization in RViz

License

MIT License