Adaptive long range vision in unstructured terrain

Ayse Naz Erkan, Raia Hadsell, Pierre Sermanet, Jan Ben, Urs Muller, Yann LeCun

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

A novel probabilistic online learning framework for autonomous off-road robot navigation is proposed. The system is purely vision-based and is particularly designed for predicting traversability in unknown or rapidly changing environments. It uses self-supervised learning to quickly adapt to novel terrains after processing a small number of frames, and it can recognize terrain elements such as paths, man-made structures, and natural obstacles at ranges up to 30 meters. The system is developed on the LAGR mobile robot platform and the performance is evaluated using multiple metrics, including ground truth.

Original languageEnglish (US)
Title of host publicationProceedings of the 2007 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2007
Pages2421-2426
Number of pages6
DOIs
StatePublished - 2007
Event2007 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2007 - San Diego, CA, United States
Duration: Oct 29 2007Nov 2 2007

Other

Other2007 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2007
CountryUnited States
CitySan Diego, CA
Period10/29/0711/2/07

Fingerprint

Supervised learning
Mobile robots
Navigation
Robots
Processing

ASJC Scopus subject areas

  • Control and Systems Engineering

Cite this

Erkan, A. N., Hadsell, R., Sermanet, P., Ben, J., Muller, U., & LeCun, Y. (2007). Adaptive long range vision in unstructured terrain. In Proceedings of the 2007 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2007 (pp. 2421-2426). [4399622] https://doi.org/10.1109/IROS.2007.4399622

Adaptive long range vision in unstructured terrain. / Erkan, Ayse Naz; Hadsell, Raia; Sermanet, Pierre; Ben, Jan; Muller, Urs; LeCun, Yann.

Proceedings of the 2007 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2007. 2007. p. 2421-2426 4399622.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Erkan, AN, Hadsell, R, Sermanet, P, Ben, J, Muller, U & LeCun, Y 2007, Adaptive long range vision in unstructured terrain. in Proceedings of the 2007 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2007., 4399622, pp. 2421-2426, 2007 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2007, San Diego, CA, United States, 10/29/07. https://doi.org/10.1109/IROS.2007.4399622
Erkan AN, Hadsell R, Sermanet P, Ben J, Muller U, LeCun Y. Adaptive long range vision in unstructured terrain. In Proceedings of the 2007 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2007. 2007. p. 2421-2426. 4399622 https://doi.org/10.1109/IROS.2007.4399622
Erkan, Ayse Naz ; Hadsell, Raia ; Sermanet, Pierre ; Ben, Jan ; Muller, Urs ; LeCun, Yann. / Adaptive long range vision in unstructured terrain. Proceedings of the 2007 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2007. 2007. pp. 2421-2426
@inproceedings{4685c9b6a54f4743b7e122039ab5fe85,
title = "Adaptive long range vision in unstructured terrain",
abstract = "A novel probabilistic online learning framework for autonomous off-road robot navigation is proposed. The system is purely vision-based and is particularly designed for predicting traversability in unknown or rapidly changing environments. It uses self-supervised learning to quickly adapt to novel terrains after processing a small number of frames, and it can recognize terrain elements such as paths, man-made structures, and natural obstacles at ranges up to 30 meters. The system is developed on the LAGR mobile robot platform and the performance is evaluated using multiple metrics, including ground truth.",
author = "Erkan, {Ayse Naz} and Raia Hadsell and Pierre Sermanet and Jan Ben and Urs Muller and Yann LeCun",
year = "2007",
doi = "10.1109/IROS.2007.4399622",
language = "English (US)",
isbn = "1424409128",
pages = "2421--2426",
booktitle = "Proceedings of the 2007 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2007",

}

TY - GEN

T1 - Adaptive long range vision in unstructured terrain

AU - Erkan, Ayse Naz

AU - Hadsell, Raia

AU - Sermanet, Pierre

AU - Ben, Jan

AU - Muller, Urs

AU - LeCun, Yann

PY - 2007

Y1 - 2007

N2 - A novel probabilistic online learning framework for autonomous off-road robot navigation is proposed. The system is purely vision-based and is particularly designed for predicting traversability in unknown or rapidly changing environments. It uses self-supervised learning to quickly adapt to novel terrains after processing a small number of frames, and it can recognize terrain elements such as paths, man-made structures, and natural obstacles at ranges up to 30 meters. The system is developed on the LAGR mobile robot platform and the performance is evaluated using multiple metrics, including ground truth.

AB - A novel probabilistic online learning framework for autonomous off-road robot navigation is proposed. The system is purely vision-based and is particularly designed for predicting traversability in unknown or rapidly changing environments. It uses self-supervised learning to quickly adapt to novel terrains after processing a small number of frames, and it can recognize terrain elements such as paths, man-made structures, and natural obstacles at ranges up to 30 meters. The system is developed on the LAGR mobile robot platform and the performance is evaluated using multiple metrics, including ground truth.

UR - http://www.scopus.com/inward/record.url?scp=51349150693&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=51349150693&partnerID=8YFLogxK

U2 - 10.1109/IROS.2007.4399622

DO - 10.1109/IROS.2007.4399622

M3 - Conference contribution

SN - 1424409128

SN - 9781424409129

SP - 2421

EP - 2426

BT - Proceedings of the 2007 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2007

ER -