Datasets

3D Pose Graph Optimization

Datasets are described in the paper below. Click on the figure to download the corresponding dataset file in g2o format. Please cite the following paper when using the datasets:

[PDF] L. Carlone, R. Tron, K. Daniilidis, and F. Dellaert. Initialization Techniques for 3D SLAM: a Survey on Rotation Estimation and its Use in Pose Graph Optimization. In IEEE Intl. Conf. on Robotics and Automation (ICRA), pages 4597-4604, 2015.

In the following paper, we replaced the covariances with isotropic ones in the datasets sphere, sphere-a, garage, cubicle, and rim:

[PDF] L. Carlone, D. M. Rosen, G. C. Calafiore, J. J. Leonard, and F. Dellaert. Lagrangian Duality in 3D SLAM: Verification Techniques and Optimal Solutions. In IEEE/RSJ Intl. Conf. on Intelligent Robots and Systems (IROS), 2015.

Sphere-a

sphere

Torus

Cube

cube

Garage

Cubicle

Rim

2D Pose Graph Optimization

Datasets are described in the paper below. Click on the figure to download the corresponding dataset file in g2o format. Please cite the following paper when using the datasets:

[PDF] L. Carlone and A. Censi. From Angular Manifolds to the Integer Lattice: Guaranteed Orientation Estimation With Application to Pose Graph Optimization. IEEE Trans. Robotics, 30(2):475-492, 2014.

INTEL

Pose graph obtained by processing the raw measurements from wheel odometry and laser range finder, acquired at the Intel Research Lab in Seattle (raw data provided by Dirk Hähnel and available here)

MIT

Pose graph obtained by processing the raw measurements from wheel odometry and laser range finder, acquired at the MIT Killian Court (raw data available here)

M3500

Manhattan world with 3500 nodes, created by Olson et al. [E. Olson, J.J. Leonard, S.J. Teller, “Fast Iterative Alignment of Pose Graphs with Poor Initial Estimates“, 2006]

M3500a

Variant of the M3500 dataset. Extra Gaussian noise with standard deviation 0.1rad is added to the relative orientation measurements

M3500b

Variant of the M3500 dataset. Extra Gaussian noise with standard deviation 0.2rad is added to the relative orientation measurements

M3500c

Variant of the M3500 dataset. Extra Gaussian noise with standard deviation 0.3rad is added to the relative orientation measurements

More datasets are described in the paper below. Click on the figure to download the corresponding dataset file in TORO format. Please cite the following paper when using the datasets:

[PDF] L. Carlone, R. Aragues, J. A. Castellanos, and B. Bona. A fast and accurate approximation for planar pose graph optimization. Intl. J. of Robotics Research, 33(7):965-987, 2014.

FR079

Pose graph obtained by processing the raw data acquired at the Freiburg Building (the relative pose measurements are also available here)

CSAIL

Pose graph obtained by processing the raw data acquired at the MIT CSAIL building (the relative pose measurements are also available here)

FRH

Pose graph obtained by processing the raw data acquired at the Freiburg University Hospital (the relative pose measurements are also available here)

INTEL

Pose graph obtained by processing the raw measurements from wheel odometry and laser range finder, acquired at the Intel Research Lab in Seattle (raw data provided by Dirk Hähnel and available here)

M3500

Manhattan world with 3500 nodes, created by Olson et al. [E. Olson, J.J. Leonard, S.J. Teller, “Fast Iterative Alignment of Pose Graphs with Poor Initial Estimates“, 2006]

M10000

Manhattan world with 10000 nodes