| |
M. Walter, R. Eustice, and J. Leonard, A Provably
Consistent Method for Imposing Exact Sparsity in Feature-based
SLAM Information Filters, In Proceedings of the 12th
International Symposium of Robotics Research (ISRR), San
Francisco, CA, USA, October 2005 |
| Abstract |
An open problem in Simultaneous Localization and Mapping (SLAM) is the
development of algorithms which scale with the size of the
environment. A few promising methods exploit the key insight that
representing the posterior in the canonical form parameterized by a
sparse information matrix provides significant advantages regarding
computational efficiency and storage requirements. Because the
information matrix is naturally dense in the case of feature-based
SLAM, additional steps are necessary to achieve sparsity. The delicate
issue then becomes one of performing this sparsification in a manner
which is consistent with the original distribution.
In this paper, we present a SLAM algorithm based in the information
form in which sparseness is preserved while maintaining consistency.
We describe an intuitive approach to controlling the population of the
information matrix by essentially ignoring a small fraction of
proprioceptive measurements whereby we track a modified version of the
posterior. In this manner, the Exactly Sparse Extended Information
Filter (ESEIF) performs exact inference, employing a model which is
conservative relative to the standard distribution. We demonstrate our
algorithm both in simulation as well as on two nonlinear datasets,
comparing it against the standard EKF as well as the Sparse Extended
Information Filter (SEIF) by Thrun \emph{et al}. The results convincingly
show that our method yields conservative estimates for the robot pose
and map which are nearly identical to those of the EKF in comparison
to the SEIF formulation which results in overconfident error bounds.
|
| Download |
[pdf] |
@inproceedings{walter05a,
AUTHOR = {M. Walter, and R. Eustice, and J. Leonard},
TITLE = {A Provably Consistent Method for Imposing Exact Sparsity in
Feature-based {SLAM} Information Filters},
BOOKTITLE = {Proceedings of the 12th International Symposium of
Robotics Research ({ISRR})},
YEAR = {2005},
MONTH = {October},
ADDRESS = {San Francisco, CA, USA}
}
|
