VT&R3 is a C++ implementation of the Teach and Repeat navigation framework. It enables a robot to be taught a network of traversable paths and then accurately repeat any network portion. VT&R3 is designed for easy adaptation to various sensor (camera/LiDAR/RaDAR/GPS) and robot combinations. The current implementation includes a feature-based visual odometry and localization pipeline that estimates the robot's motion from stereo camera images and a point-cloud-based odometry and localization pipeline for LiDAR sensors.
The following articles will help you get started with VT&R3:
- Installation Guide
- Launching VT&R3
- Setting Up VT&R3 with a New Sensor or Robot
- VT&R3 Sample Datasets
More information can be found in the wiki page.
Please cite the following paper when using VT&R3 for your research:
@article{paul2010vtr,
author = {Furgale, Paul and Barfoot, Timothy D.},
title = {Visual teach and repeat for long-range rover autonomy},
journal = {Journal of Field Robotics},
year = {2010},
doi = {https://doi.org/10.1002/rob.20342}
}- Multi-Experience Localization
@inproceedings{michael2016mel, author = {Paton, Michael and MacTavish, Kirk and Warren, Michael and Barfoot, Timothy D.}, booktitle = {2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)}, title = {Bridging the appearance gap: Multi-experience localization for long-term visual teach and repeat}, year={2016}, doi={10.1109/IROS.2016.7759303} }