Incremental Pose Graph Optimization


In previous work, we showed how the SLAM problem can be cast as a non-linear optimization problem and presented a solution similar to stochastic gradient descent. We now extend our previous batch solution, allowing new constraints to be added.


Spatially-Adaptive Learning Rates for Online Incremental SLAM (PDF) (1.6 MB)

Learning rates tree data structure source code (

@inproceedings{ Olson-RSS-07,
	AUTHOR    = {E. Olson and J. Leonard and S. Teller},
	TITLE     = {Spatially-Adaptive Learning Rates for Online Incremental SLAM},
	BOOKTITLE = {Proceedings of Robotics: Science and Systems},
	YEAR      = {2007},
	ADDRESS   = {Atlanta, GA, USA},
	MONTH     = {June}


Processing the Freiburg data set (XVID)

Robotics, Vision, and Sensor Networks Group
32 Vassar Street, 32-33x
Cambridge, MA 02139
Tel: 617-253-6583
Fax: 617-258-7413
Logo of the MIT Computer Science and Artificial Intelligence Laboratory