- Purpose: Coordinate transformations are essential in Autonomous Mobile Robots (AMR) to relate different sensor data and robot positions across various reference frames.
- Key Frames: Include Map, World, Odometry (global); Base Link, Laser, Camera (robot-specific).
- ROS TF: Provides tools for managing and applying transformations between frames, crucial for integrating sensor data and robot control.
- Static Transformations: Fixed relationships (e.g., sensor to base), published on
/tf_static
, useStaticTransformBroadcaster
. - Dynamic Transformations: Changing relationships (e.g., robot to starting point), published on
/tf
, useTransformBroadcaster
. - Key Points: Both use TF broadcasters, include translation and rotation, listeners retrieve data, TF tree represents frame hierarchy.
- Robot State Publisher: Broadcasts robot state to tf library, uses URDF to compute and publish 3D poses of links.
- Joint State Publisher: Publishes JointState messages for non-fixed joints, sources include GUI, subscribed messages, or default values.
- Key Points: Automates TF publishing from URDF, handles non-fixed joints, useful for testing and visualization, real-world applications use controller_manager.