Difference between revisions of "RTAB-Map"
| Line 19: | Line 19: | ||
''Note that these results (more recent) may differ from those in the video...'' | ''Note that these results (more recent) may differ from those in the video...'' | ||
| − | Summary of the loop closures detected on UdeS1Hz data set : | + | Figure 1: Summary of the loop closures detected on UdeS1Hz data set : |
* Green : Loop closures detected | * Green : Loop closures detected | ||
* Yellow : Loop closures rejected | * Yellow : Loop closures rejected | ||
* Red : Unable to detect a loop closure because old places could not be retrieved | * Red : Unable to detect a loop closure because old places could not be retrieved | ||
| − | |||
| − | |||
| − | |||
| − | Processing time for each image acquired (real-time limit fixed to 700 ms for an image rate of 1 Hz) | + | Figure 2: Processing time for each image acquired (real-time limit fixed to 700 ms for an image rate of 1 Hz) |
| − | |||
| − | |||
| − | |||
| − | Precision-Recall (43% recall at 100% precision) | + | Figure 3: Precision-Recall (43% recall at 100% precision) |
| − | + | ||
| − | [[File:RTAB-Map_RecallResults.png| | + | [[File:RTAB-Map_LoopClosureMapResults.png|250px]] [[File:RTAB-Map_LoopClosureTimeResults.png|250px]] [[File:RTAB-Map_RecallResults.png|250px]] |
| − | |||
</english> | </english> | ||
| Line 42: | Line 35: | ||
''À noter que les résultats (plus récents) présentés ici peuvent différer de ceux dans le vidéo...'' | ''À noter que les résultats (plus récents) présentés ici peuvent différer de ceux dans le vidéo...'' | ||
| − | Sommaire des détection de boucles sur l'ensemble de données UdeS1Hz : | + | Figure 1: Sommaire des détection de boucles sur l'ensemble de données UdeS1Hz : |
* Vert : Fermetures de boucle acceptées | * Vert : Fermetures de boucle acceptées | ||
* Jaune : Fermetures de boucle rejetées | * Jaune : Fermetures de boucle rejetées | ||
* Rouge : Impossibilité de détecter une fermeture de boucle car les anciens endroits n'ont pu être remémorisés | * Rouge : Impossibilité de détecter une fermeture de boucle car les anciens endroits n'ont pu être remémorisés | ||
| − | |||
| − | |||
| − | |||
| − | Temps d'exécution pour chaque itération (limite temps réel fixée à 700 ms pour un temps d'acquisition de 1 seconde) | + | Figure 2: Temps d'exécution pour chaque itération (limite temps réel fixée à 700 ms pour un temps d'acquisition de 1 seconde) |
| − | + | ||
| − | [[File:RTAB-Map_LoopClosureTimeResults.png| | + | Figure 3: Precision-Recall (43% recall à 100% precision) |
| − | + | ||
| + | [[File:RTAB-Map_LoopClosureMapResults.png|250px]] [[File:RTAB-Map_LoopClosureTimeResults.png|250px]] [[File:RTAB-Map_RecallResults.png|250px]] | ||
| + | |||
</french> | </french> | ||
Revision as of 20:33, 12 April 2011
RTAB-Map : Real-Time Appearance-Based Mapping
Description[edit]
Loop closure detection is the process involved when trying to find a match between the current and a previously visited locations in SLAM (Simultaneous Localization And Mapping). Over time, the amount of time required to process new observations increases with the size of the internal map, which may affect real-time processing. RTAB-Map is a novel real-time loop closure detection approach for large-scale and long-term SLAM. Our approach is based on efficient memory management to keep computation time for each new observation under a fixed time limit, thus respecting real-time limit for long-term operation. Results demonstrate the approach's adaptability and scalability using one custom data set and four standard data sets.
Results[edit]
Note that these results (more recent) may differ from those in the video...
Figure 1: Summary of the loop closures detected on UdeS1Hz data set :
- Green : Loop closures detected
- Yellow : Loop closures rejected
- Red : Unable to detect a loop closure because old places could not be retrieved
Figure 2: Processing time for each image acquired (real-time limit fixed to 700 ms for an image rate of 1 Hz)
Figure 3: Precision-Recall (43% recall at 100% precision)
Source code[edit]
The code was tested on Windows (Xp, 7), Mac OS X 10.6 and Ubuntu 10.4LTS.
- Standalone application and libraries : rtabmap
- ROS packages : rtabmap-ros-pkg
Data sets[edit]
UdeS1Hz samples[edit]
Downloads[edit]
Community data sets from other loop closure detection approaches :
- Angeli et al. : Lip6Indoor and Lip6Outdoor
- Cummins et al. (FAB-MAP) : NewCollege and CityCentre
Publications
Labbé, M., Michaud, F. (2011), “Memory management approach for real-time appearance-based loop closure detection”, To appear in IEEE Transactions on Robotics.
