General Details
===============

You can use RKO-LIO in two primary ways:

1. **Offline** using the Python application ``rko_lio`` for analyzing pre-recorded data.
2. **Online** on a robot via our ROS packages.

The ROS-based system is the recommended approach for using RKO-LIO.
Online or even offline, as we provide additional nodes to process ROS bags directly as well (see *offline mode* in :doc:`ROS Usage <../ros>`).
It offers **better performance** due to more efficient handling of incoming data streams.
If possible, prefer using the ROS version.
The Python version is intended mainly for convenience.

For odometry-specific parameter configuration, refer to :doc:`Configuring rko_lio <config>`.


.. toctree::
   :maxdepth: 1
   :caption: Contents:

   Sensor data <data>
   Configuring rko_lio <config>
   Core dependencies <core_dependencies>

For a short demo video of RKO-LIO, you can check this YouTube video:

.. figure:: ../../_static/odometry_video_thumbnail.png
   :target: https://www.youtube.com/watch?v=NNpzXdf9XmU