libRSF - A Robust Sensor Fusion Library

The libRSF is an open source C++ library that provides the basic components for robust sensor fusion. It can be used to describe an estimation problem as a factor graph and solves it with least squares, powered by the Ceres Solver. More information can be found under libRSF - A Robust Sensor Fusion Library.

Main features are:

A sliding window filter for online applications, including marginalization.
A set of predefined cost functions for various localization problems.
Several robust error models for non-Gaussian problems, including self-tuning Gaussian mixtures.

Build and Test Status

Platform	Status
Ubuntu 22.04
Ubuntu 20.04

Installation

The libRSF is a CMake project that requires the installation of several dependencies. For convenience, we provide a simple bash script that installs required packages. It is tested only for Ubuntu 20.04/22.04:

  git clone https://github.com/TUC-ProAut/libRSF.git
  cd libRSF
  bash InstallDependencies.bash

Alternatively, you can install them by your own:

CMake (>= 3.5)
```
sudo apt-get install cmake
```

Eigen (>= 3.3.5)

Only for Ubuntu < 20.04, you have to install a current version of Eigen locally.

mkdir -p externals/install
git submodule update --init externals/eigen

cd externals/eigen
mkdir -p build && cd build
cmake -DCMAKE_INSTALL_PREFIX=../../install/ ..
make install
cd ../../..

For Ubuntu >= 20.04 you can install Eigen straight-forward.

sudo apt-get install libeigen3-dev

Ceres (>= 2.0) and its dependencies

sudo apt-get install libgoogle-glog-dev
sudo apt-get libgflags-dev
sudo apt-get install libatlas-base-dev
sudo apt-get install libsuitesparse-dev

mkdir -p externals/install  
git submodule update --init externals/ceres-solver

cd externals/ceres-solver
mkdir build && cd build
cmake -DEigen3_DIR=../install/share/eigen3/cmake -DCMAKE_INSTALL_PREFIX=../../install/ ..
make all -j$(getconf _NPROCESSORS_ONLN)
make install
cd ../..

yaml-cpp
```
sudo apt-get install libyaml-cpp-dev
```
GeographicLib
```
sudo apt-get install libgeographic-dev
```

The library and its applications can be build following this instructions:

  git clone https://github.com/TUC-ProAut/libRSF.git
  cd libRSF
  mkdir build && cd build
  cmake ..
  make all -j$(getconf _NPROCESSORS_ONLN)

You can install the libRSF using:

  make install

And remove it using:

  make uninstall

Usage

After building the library, some applications are provided which correspond directly to a publication. The following pages give you an overview, how to use them or how to build a custom application using the libRSF:

Additional Information

Citation

If you use this library for academic work, please cite it using the following BibTeX reference:

  @Misc{libRSF,
   author       = {Tim Pfeifer and Others},
   title        = {libRSF},
   howpublished = {\url{https://github.com/TUC-ProAut/libRSF}}
  }

This library also contains the implementation of [1-3]. Further references will be added with additional content.

[1] Tim Pfeifer and Peter Protzel, Expectation-Maximization for Adaptive Mixture Models in Graph Optimization, Proc. of Intl. Conf. on Robotics and Automation (ICRA), 2019, DOI: 10.1109/ICRA.2019.8793601

[2] Tim Pfeifer and Peter Protzel, Incrementally learned Mixture Models for GNSS Localization, Proc. of Intelligent Vehicles Symposium (IV), 2019, DOI: 10.1109/IVS.2019.8813847

[3] Tim Pfeifer and Sven Lange and Peter Protzel, Advancing Mixture Models for Least Squares Optimization, Robotics and Automation Letters (RA-L), 2021, DOI: 10.1109/LRA.2021.3067307

License

This work is released under the GNU General Public License version 3.