Kimera-RPGO: Open-Source Robust Pose Graph Optimization

Authors: Yun Chang, Luca Carlone, Marcus Abate, Antoni Rosinol

Publications

We kindly ask to cite our paper if you find this library useful:

• A. Rosinol, M. Abate, Y. Chang, L. Carlone, Kimera: an Open-Source Library for Real-Time Metric-Semantic Localization and Mapping. IEEE Intl. Conf. on Robotics and Automation (ICRA), 2020. arXiv:1910.02490.

@InProceedings{Rosinol20icra-Kimera,
  title = {Kimera: an Open-Source Library for Real-Time Metric-Semantic Localization and Mapping},
  author = {Rosinol, Antoni and Abate, Marcus and Chang, Yun and Carlone, Luca},
  year = {2020},
  booktitle = {IEEE Intl. Conf. on Robotics and Automation (ICRA)},
  url = {https://github.com/MIT-SPARK/Kimera},
  pdf = {https://arxiv.org/pdf/1910.02490.pdf}
}

Related Publications

Outlier Rejection based on:

• J. Mangelson, D. Dominic, R. Eustice, and R. Vasudevan. Pairwise Consistent Measurement Set Maximization for Robust Multi-robot Map Merging. IEEE Intl. Conf. on Robotics and Automation (ICRA), 2018.

Maximum Clique Finder used with slight adaptations:

• B. Pattabiraman, M. Patwary, A. Gebremedhin, W. Liao, and A. Choudhary. Fast Algorithms for the Maximum Clique Problem on Massive Graphs with Applications to Overlapping Community Detection, Internet Mathematics, 2014.

• B. Pattabiraman, M. Patwary, A. Gebremedhin, W. Liao, and A. Choudhary. Fast Algorithms for the Maximum Clique Problem on Massive Sparse Graphs, In Proceedings of the 10th Workshop on Algorithms and Models for the Web Graph, Cambridge, MA, Lecture Notes in Computer Science, Springer, vol. 8305, pp. 156-169, 2013.

Dependencies

GTSAM (Note that the BUILD_WITH_MARCH_NATIVE flag caused some problems on some machines. )

Clone GTSAM to your preferred location:

git clone [email protected]:borglab/gtsam.git

Build

cd gtsam
mkdir build
cd build
cmake .. -DGTSAM_POSE3_EXPMAP=ON -DGTSAM_ROT3_EXPMAP=ON
sudo make install

Note: The reason why we need EXPMAP is for the correct calculation of Jacobians. Enabling this and the #define SLOW_BUT_CORRECT_BETWEENFACTOR before any #include <gtsam/slam/BetweenFactor.h> are both important. Otherwise the default are some identity approximations for rotations, which works for some cases but fails for cases with manual loop closures, or artifacts. Note that sudo make check in gtsam will partially fail because some unittests assume the EXPMAP flags to be off.

Build

git clone [email protected]:MIT-SPARK/Kimera-RPGO.git
cd Kimera-RPGO
mkdir build
cd build
cmake ..
make

Usage

This repository can be used as an optimization backend. A sample setup looks something like below. The default solver is LM.

// Set up
// set up KimeraRPGO solver
RobustSolverParams params;
params.setPcm3DParams(0.0, 10.0, Verbosity::QUIET);
// Verbosity levels are QUIET, UPDATE, and, VERBOSE in order of increasing number of messages (the default is UPDATE)
// For 2D params.setPcm2DParams(0.0, 10.0); Have been tested

// To use GaussNewton instead of LM: params.solver = Solver::GN;

std::unique_ptr<RobustSolver> pgo = KimeraRPGO::make_unique<RobustSolver>(params);
//...
//...

// Run
pgo->update(new_factor, new_values);

This can also be used as a standalone experimental tool. A read g2o function can be found in examples.

# for 2D:
./RpgoReadG2o 2d <g2o-file> <odom-check-threshold> <pcm-threshold> <optional:folder-to-save-g2o> <optional:v to toggle verbosity>

# for 3D
./RpgoReadG2o 3d <g2o-file> <odom-check-threshold> <pcm-threshold> <optional:folder-to-save-g2o> <optional:v to toggle verbosity>

Example, do ./RpgoReadG2o 3d /home/user/Desktop/in.g2o 1.0 1.0 /home/user/Desktop/out/ v

Example

// set up KimeraRPGO solver
RobustSolverParams params;
params.setPcm3DParams(<translation_threshold>, <rotation_threshold>);
std::vector<char> special_symbs{'l', 'u'}; // for landmarks
params.specialSymbols = special_symbs;

std::unique_ptr<RobustSolver> pgo = KimeraRPGO::make_unique<RobustSolver>(params); // initiate pgo solver

// When using it normally
pgo->update(new_factor, new_values);

// load a graph (assuming pgo has been reset)
pgo->loadGraph(nfg, values, prior_factor);
pgo->loadGraph(nfg, values, key0); // can just load the first key

// add a graph (ex. already have graph or robot a and adding graph of robot b)
pgo->addGraph(nfg, values, between_factor);


/////// To not use outlier rejection, set up like
RobustSolverParams params;
params.setNoRejection();

std::unique_ptr<RobustSolver> pgo = KimeraRPGO::make_unique<RobustSolver>(params); // initiate pgo solver

BSD License

Kimera-RPGO is open source under the BSD license, see the LICENSE.BSD file.