welcome: please sign in
location: Diff for "rtslam"
Differences between revisions 5 and 31 (spanning 26 versions)
⇤ ← Revision 5 as of 2010-12-13 21:34:33 →
Size: 1686
Editor: cyril42e-wifi
Comment: +provides
← Revision 31 as of 2014-07-24 13:43:17 → ⇥
Size: 3712
Editor: croussil
Comment: add alias before authors to ask people to use it
Deletions are marked like this. Additions are marked like this.
Line 5: Line 5:

<!> '''Disclaimer:''' RT-SLAM is currently in development and has no official release. It is provided as-is, there is no standalone installation, there is some code to adapt for your own use, and it is generally not advised (yet) to people who do not want to get their hands dirty.
Line 11: Line 8:
 * '''genericity''': sensor models, landmark types, landmark models, reparametrization from one landmark model to another one, estimating biases;
 * '''speed''': real time at 60 fps, VGA, gray level;
 * '''flexibility''': different estimation/image processing sequencing (active search), independent base brick for a hierarchical multimap and multirobots architecture;
 * '''robustness''' : near-optimal repartition of landmarks, data association errors detection (gating, ransac);
 * '''developer-friendly''' : visualization tools (2D and 3D), offline replay step by step, logs, simulation.		  * '''genericity''': for sensor models, landmark types, landmark models, landmarks reparametrization, biases estimation;
 * '''speed''': real time at 60 fps (VGA, gray level) on a decent machine (at least one Core2 core at 2.2GHz)
 * '''flexibility''': different estimation/image processing sequencing strategies (active search), independent base brick for a hierarchical multimap and multirobots architecture;
 * '''robustness''': near-optimal repartition of landmarks, data association errors detection (gating, ransac);
 * '''developer-friendly''': visualization tools (2D and 3D), offline replay step by step, logs, simulation.
 * '''open source''': in C++ for Linux and MacOS
Line 18: Line 16:
 * Anchored Homogeneous Points (Inverse Depth) that can be reparametrized into Euclidean Points
 * Pinhole cameras
 * Constant Velocity prediction model
 * Active search and 1-point Ransac for data association		  * '''Landmarks''': Anchored Homogeneous Points (Inverse Depth) that can be reparametrized into Euclidean Points;
 * '''Sensors''': Pinhole cameras;
 * '''Prediction''': Constant velocity model, inertial sensor;
 * '''Data association''': Active search, 1-point Ransac, and mixed strategies.
Line 26: Line 24:
  * [[rtslam/Installation|Installation procedure]]   * [[rtslam/Installation|Installation guide]]
Line 28: Line 26:
  * [[http://homepages.laas.fr/croussil/doc/jafar/group__rtslam.html|Developer reference]]
   * [[http://homepages.laas.fr/croussil/doc/jafar/group__rtslam.html|Developer guide]] (Doxygen documentation)
  * [[http://trac.laas.fr/git/rtslam-jafar|Git web interface]] (source code browsing)
  * Contact: rtslam-users@laas.fr (please always '''reply all''' to messages to include the mailing list; if you are interested you can subscribe to this alias [[https://sympa.laas.fr/sympa/info/rtslam-users|there]])
Line 33: Line 32:
||<tablestyle="width:100%;text-align:center;" style="border:0px"> {{attachment:rtslam3D.png|3D view|width=400px}}||<style="border:0px"> {{attachment:rtslam2D.png|2D view|width=400px}} ||

 * [[http://homepages.laas.fr/croussil/videos/2010-11-29_rtslam_grande-salle_q25.avi|Demo 1]] (v.a, 1'33, 23Mo, H264 AVI): hand held alone camera at 60 fps indoor, with several loop closures.
 * [[http://homepages.laas.fr/croussil/videos/2011-02-15_rtslam-inertial-high-dyn.avi|Demo 2]] (v.b, 1'06, 20Mo, H264 AVI): hand held IMU+camera at 50 fps indoor, with very high dynamic.
 * [[http://homepages.laas.fr/croussil/videos/2011-06-09_rtslam-inertial-robot-grass.avi|Demo 3]] (v.a, 1'25, 27Mo, H264 AVI): IMU+camera on a rover robot on grass, with short term memory (lost landmarks are removed, no loop closure)
 * [[/Material|Additional videos referenced in papers]]

== Papers ==

  * __RT-SLAM: A Generic and Real-Time Visual SLAM Implementation__. Cyril Roussillon, Aurélien Gonzalez, Joan Solà, Jean-Marie Codol, Nicolas Mansard, Simon Lacroix, and Michel Devy. ICVS 2011, volume 6962 of Lecture Notes in Computer Science, page 31-40. Springer. [[http://homepages.laas.fr/croussil/publications/2011_ICVS_rtslam.pdf|Download PDF]]
  * __High rate-localization for high-speed all-terrain robots__. Cyril Roussillon, Simon Lacroix. CCCA 2012. [[http://homepages.laas.fr/croussil/publications/2012_CCCA_rtslam.pdf|Download PDF]]
Line 38: Line 48:
 * Inertial and Odometry Slam  * Odometry Slam
Line 40: Line 50:
 * Segment Slam  * Segments Slam

== Authors ==

 * Contact for general questions: rtslam-users@laas.fr
 * [[http://joansola.eu|Joan Solà]]
 * [[http://pro.cyril-roussillon.fr|Cyril Roussillon]]
 * [[http://homepages.laas.fr/agonzale|Aurélien Gonzalez]]
 * [[http://homepages.laas.fr/nmansard|Nicolas Mansard]]
 * [[http://homepages.laas.fr/jmcodol|Jean-Marie Codol]]
 * [[http://homepages.laas.fr/dmarquez|David Marquez]]

RT-SLAM

RT-SLAM stands for Real Time SLAM (Simultaneous Localization And Mapping).

Presentation

RT-SLAM is a fast Slam library and test framework based on EKF. Its main qualities are:

  • genericity: for sensor models, landmark types, landmark models, landmarks reparametrization, biases estimation;

  • speed: real time at 60 fps (VGA, gray level) on a decent machine (at least one Core2 core at 2.2GHz)

  • flexibility: different estimation/image processing sequencing strategies (active search), independent base brick for a hierarchical multimap and multirobots architecture;

  • robustness: near-optimal repartition of landmarks, data association errors detection (gating, ransac);

  • developer-friendly: visualization tools (2D and 3D), offline replay step by step, logs, simulation.

  • open source: in C++ for Linux and MacOS

For now it provides:

  • Landmarks: Anchored Homogeneous Points (Inverse Depth) that can be reparametrized into Euclidean Points;

  • Sensors: Pinhole cameras;

  • Prediction: Constant velocity model, inertial sensor;

  • Data association: Active search, 1-point Ransac, and mixed strategies.

Documentation

Screenshots and videos

3D view

2D view

  • Demo 1 (v.a, 1'33, 23Mo, H264 AVI): hand held alone camera at 60 fps indoor, with several loop closures.

  • Demo 2 (v.b, 1'06, 20Mo, H264 AVI): hand held IMU+camera at 50 fps indoor, with very high dynamic.

  • Demo 3 (v.a, 1'25, 27Mo, H264 AVI): IMU+camera on a rover robot on grass, with short term memory (lost landmarks are removed, no loop closure)

  • Additional videos referenced in papers

Papers

  • RT-SLAM: A Generic and Real-Time Visual SLAM Implementation. Cyril Roussillon, Aurélien Gonzalez, Joan Solà, Jean-Marie Codol, Nicolas Mansard, Simon Lacroix, and Michel Devy. ICVS 2011, volume 6962 of Lecture Notes in Computer Science, page 31-40. Springer. Download PDF

  • High rate-localization for high-speed all-terrain robots. Cyril Roussillon, Simon Lacroix. CCCA 2012. Download PDF

Roadmap

  • Stabilize and make a release
  • Odometry Slam
  • Multimap Slam
  • Segments Slam

Authors

OpenrobotsWiki: rtslam (last edited 2014-07-24 13:43:17 by croussil)