CaltechAUTHORS
  A Caltech Library Service

Unsupervised Learning for Subterranean Junction Recognition Based on 2D Point Cloud

Sharif Mansouri, Sina and Pourkamali-Anaraki, Farhad and Castano, Miguel and Agha-Mohammadi, Ali-Akbar and Burdick, Joel and Nikolakopoulos, George (2020) Unsupervised Learning for Subterranean Junction Recognition Based on 2D Point Cloud. In: 2020 28th Mediterranean Conference on Control and Automation (MED). IEEE , Piscataway, NJ, pp. 802-807. ISBN 9781728157429. https://resolver.caltech.edu/CaltechAUTHORS:20200911-083206534

[img] PDF - Submitted Version
See Usage Policy.

1328Kb

Use this Persistent URL to link to this item: https://resolver.caltech.edu/CaltechAUTHORS:20200911-083206534

Abstract

This article proposes a novel unsupervised learning framework for detecting the number of tunnel junctions in subterranean environments based on acquired 2D point clouds. The implementation of the framework provides valuable information for high level mission planners to navigate an aerial platform in unknown areas or robot homing missions. The framework utilizes spectral clustering, which is capable of uncovering hidden structures from connected data points lying on non-linear manifolds. The spectral clustering algorithm computes a spectral embedding of the original 2D point cloud by utilizing the eigen decomposition of a matrix that is derived from the pairwise similarities of these points. We validate the developed framework using multiple data-sets, collected from multiple realistic simulations, as well as from real flights in underground environments, demonstrating the performance and merits of the proposed methodology.


Item Type:Book Section
Related URLs:
URLURL TypeDescription
https://doi.org/10.1109/med48518.2020.9183337DOIArticle
https://arxiv.org/abs/2006.04225arXivDiscussion Paper
ORCID:
AuthorORCID
Sharif Mansouri, Sina0000-0001-7631-002X
Pourkamali-Anaraki, Farhad0000-0003-4078-1676
Agha-Mohammadi, Ali-Akbar0000-0001-5509-1841
Nikolakopoulos, George0000-0003-0126-1897
Additional Information:© 2020 IEEE. This work has been partially funded by the European Unions Horizon 2020 Research and Innovation Programme under the Grant Agreement No. 730302 SIMS. Funding from Vinnova in the project ‘AI Factory for Railway’ is also acknowledged.
Funders:
Funding AgencyGrant Number
European Research Council (ERC)730302
VinnovaUNSPECIFIED
Record Number:CaltechAUTHORS:20200911-083206534
Persistent URL:https://resolver.caltech.edu/CaltechAUTHORS:20200911-083206534
Official Citation:S. S. Mansouri, F. Pourkamali-Anaraki, M. Castano, A. Agha-Mohammadi, J. Burdick and G. Nikolakopoulos, "Unsupervised Learning for Subterranean Junction Recognition Based on 2D Point Cloud," 2020 28th Mediterranean Conference on Control and Automation (MED), Saint-Raphaël, France, 2020, pp. 802-807, doi: 10.1109/MED48518.2020.9183337
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:105317
Collection:CaltechAUTHORS
Deposited By: Tony Diaz
Deposited On:11 Sep 2020 16:03
Last Modified:11 Sep 2020 16:03

Repository Staff Only: item control page