This is a restructure/rewrite of the amazing work of Andreu Huguet. His original work canbe found at LIMO-Velo. More information on why this fork exists can be found in the Motivation section.

Visualization of the algorithm with delta = 0.01 (100Hz)

Designed for easy modifying via modular and easy to understand code. Relying upon HKU-Mars's IKFoM and ikd-Tree open-source libraries. Based also on their FAST_LIO2.

Common working speeds are 20m/s in straights and 100deg/s in the turns.

Comparison of cones under racing speeds running all algorithms in real-time, except for LIO-SAM (-r 0.5). It failed otherwise.

The easiest way to experiment is through the Python interface.

For convinience and the ability to easily experiment with the algorithm both offline and in real time, a Python interface is provided. This interface is achived through the use of the pybind11 library. Essentially, the C++ code is compiled into a Python module that can be imported and used in Python, ensuring that the performance of the C++ code is maintained and the same behaviour is observed.

To install the Python interface, run the following commands:

pip install limovelo

The Python interface is designed to be as simple as possible. The following is an example of how to use the Python interface:

import limovelo

# Create the LIMO-Velo object
limo = limovelo.LIMO_Velo()

# Load the configuration file
limo.load_config("config.yaml")

# Load the LiDAR data
lidar_data = limovelo.LidarData()
lidar_data.load("lidar_data.bin")

# Load the IMU data
imu_data = limovelo.ImuData()
imu_data.load("imu_data.bin")

The code is made as a C++ library, and can be used as such. Refer to the ROS2 implementation for an example of how to use the library.

The ROS2 implementation is a wrapper around the C++ library, and is designed to be used in a ROS2 environment. The following is an example of how to use the ROS2 implementation:

colcon build
ros2 run limovelo limovelonode

• LIMO-Velo: The official repository of LIMO-Velo from which this fork is based.
• IKFoM: Iterated Kalman Filters on Manifolds
• ikd-Tree: Incremental KD-Tree for Robotic Applications
• FAST-LIO2: Fast and Direct LIO SLAM