kf_cpp

ROS 2 C++ implementation of Extended Kalman filter (EKF) and Unscented Kalman filter (UKF) with demo nodes.

Table of Contents

• kf_cpp
  • Table of Contents
  • Installation
    • Prerequisites
      • Dependencies from CMakeLists.txt
  • KF v.s. EKF v.s. UKF
  • Bibliography
  • Acknowledgements
  • Contact

Installation

This repository is ROS 2-based and utilises several dependencies that are common in robotics applications. Follow the instructions below to build the package in a ROS 2 environment.

Prerequisites

Make sure you have the following installed:

• ROS 2 humble: Installed and properly sourced. Follow the ROS 2 installation guide if you haven't set it up yet.
• C++17 Compiler: Ensure your compiler supports C++17.
• Dependencies: The package depends on several libraries and ROS 2 packages.

Dependencies from CMakeLists.txt

• Eigen3
• ROS 2 packages:
  • rclcpp

Clone this repo into your ROS 2 workspace, install dependencies, and build pkg:

cd ~/ros2_ws/src && git clone https://github.com/wei-hsuan-cheng/kf_cpp.git

cd ~/ros2_ws && rosdep update && rosdep install --from-paths src --ignore-src -r -y

cd ~/ros2_ws && colcon build --packages-select kf_cpp && . install/setup.bash

Run the ROS 2 node for test:

cd ~/ros2_ws && . install/setup.bash
ros2 run kf_cpp ukf_ori_sim

KF v.s. EKF v.s. UKF

The table below compares the process of each filter in each iteration.

Bibliography

• EKF:
  • [1]
• UKF:

  • [1] Rudolph Van der Merwe. "Sigma-Point Kalman Filters for Probabilistic Inference in Dynamic State-Space Models" dissertation (2004).

  • [2] Simon J. Julier. "The Scaled Unscented Transformation". Proceedings of the American Control Conference 6. IEEE. (2002)

  • [3] http://www.esdradar.com/brochures/Compact%20Tracking%2037250X.pdf

  • [4] Julier, Simon J.; Uhlmann, Jeffrey "A New Extension of the Kalman Filter to Nonlinear Systems". Proc. SPIE 3068, Signal Processing, Sensor Fusion, and Target Recognition VI, 182 (July 28, 1997)

  • [5] Cholesky decomposition. Wikipedia. http://en.wikipedia.org/wiki/Cholesky_decomposition

Acknowledgements

• Kalman-and-Bayesian-Filters-in-Python: The EKF algorithm is an C++ implementation adapted from the Kalman-and-Bayesian-Filters-in-Python.
• hdl-localization-ROS2: The UKF algorithm is adapted from the hdl-localization-ROS2 codebase.
• Robot Math Utils: Some utilities from Robot Math Utils are used.
• Eigen Library: This library heavily relies on the Eigen library for linear algebra operations.

Contact

• Author: Wei-Hsuan Cheng ([email protected])
• Homepage: wei-hsuan-cheng
• GitHub: wei-hsuan-cheng