EchoSLAM: Simultaneous Localization and Mapping with Acoustic Echoes

We address the problem of jointly localizing a robot in an unknown room and estimating the room geometry from echoes. Unlike earlier work using echoes, we assume a completely autonomous setup with (near) collocated microphone and the acoustic source. We first introduce a simple, easy to analyze estimator, and prove that the sequence of room and trajectory estimates converges to the true values. Next, we approach the problem from a Bayesian point of view, and propose a more general solution which does not require any assumptions on motion and measurement model of the robot. In addition to theoretical analysis, we validate both estimators numerically.


Published in:
2016 Ieee International Conference On Acoustics, Speech And Signal Processing Proceedings, 11-15
Presented at:
41st IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2016), Shanghai, China, 20-25 March 2016
Year:
2016
Publisher:
New York, Ieee
ISSN:
1520-6149
ISBN:
978-1-4799-9988-0
Keywords:
Laboratories:




 Record created 2016-01-25, last modified 2018-03-17

Preprint:
Download fulltext
PDF

Rate this document:

Rate this document:
1
2
3
 
(Not yet reviewed)