Deep belief net learning in a long-range vision system for autonomous off-road driving

Raia Hadsell, Ayse Erkan, Pierre Sermanet, Marco Scoffier, Urs Muller, Yann LeCun

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

Abstract

We present a learning-based approach for longrange vision that is able to accurately classify complex terrain at distances up to the horizon, thus allowing high-level strategic planning. A deep belief network is trained with unsupervised data and a reconstruction criterion to extract features from an input image, and the features are used to train a realtime classifier to predict traversability. The online supervision is given by a stereo module that provides robust labels for nearby areas up to 12 meters distant. The approach was developed and tested on the LAGR mobile robot.

Original languageEnglish (US)
Title of host publication2008 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS
Pages628-633
Number of pages6
DOIs
StatePublished - 2008
Event2008 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS - Nice, France
Duration: Sep 22 2008Sep 26 2008

Other

Other2008 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS
CountryFrance
CityNice
Period9/22/089/26/08

Fingerprint

Strategic planning
Bayesian networks
Mobile robots
Labels
Classifiers

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Vision and Pattern Recognition
  • Control and Systems Engineering
  • Electrical and Electronic Engineering

Cite this

Hadsell, R., Erkan, A., Sermanet, P., Scoffier, M., Muller, U., & LeCun, Y. (2008). Deep belief net learning in a long-range vision system for autonomous off-road driving. In 2008 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS (pp. 628-633). [4651217] https://doi.org/10.1109/IROS.2008.4651217

Deep belief net learning in a long-range vision system for autonomous off-road driving. / Hadsell, Raia; Erkan, Ayse; Sermanet, Pierre; Scoffier, Marco; Muller, Urs; LeCun, Yann.

2008 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS. 2008. p. 628-633 4651217.

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

Hadsell, R, Erkan, A, Sermanet, P, Scoffier, M, Muller, U & LeCun, Y 2008, Deep belief net learning in a long-range vision system for autonomous off-road driving. in 2008 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS., 4651217, pp. 628-633, 2008 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS, Nice, France, 9/22/08. https://doi.org/10.1109/IROS.2008.4651217
Hadsell R, Erkan A, Sermanet P, Scoffier M, Muller U, LeCun Y. Deep belief net learning in a long-range vision system for autonomous off-road driving. In 2008 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS. 2008. p. 628-633. 4651217 https://doi.org/10.1109/IROS.2008.4651217
Hadsell, Raia ; Erkan, Ayse ; Sermanet, Pierre ; Scoffier, Marco ; Muller, Urs ; LeCun, Yann. / Deep belief net learning in a long-range vision system for autonomous off-road driving. 2008 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS. 2008. pp. 628-633
@inproceedings{8dcab34f79d34c85b50577bbb625976c,
title = "Deep belief net learning in a long-range vision system for autonomous off-road driving",
abstract = "We present a learning-based approach for longrange vision that is able to accurately classify complex terrain at distances up to the horizon, thus allowing high-level strategic planning. A deep belief network is trained with unsupervised data and a reconstruction criterion to extract features from an input image, and the features are used to train a realtime classifier to predict traversability. The online supervision is given by a stereo module that provides robust labels for nearby areas up to 12 meters distant. The approach was developed and tested on the LAGR mobile robot.",
author = "Raia Hadsell and Ayse Erkan and Pierre Sermanet and Marco Scoffier and Urs Muller and Yann LeCun",
year = "2008",
doi = "10.1109/IROS.2008.4651217",
language = "English (US)",
isbn = "9781424420582",
pages = "628--633",
booktitle = "2008 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS",

}

TY - GEN

T1 - Deep belief net learning in a long-range vision system for autonomous off-road driving

AU - Hadsell, Raia

AU - Erkan, Ayse

AU - Sermanet, Pierre

AU - Scoffier, Marco

AU - Muller, Urs

AU - LeCun, Yann

PY - 2008

Y1 - 2008

N2 - We present a learning-based approach for longrange vision that is able to accurately classify complex terrain at distances up to the horizon, thus allowing high-level strategic planning. A deep belief network is trained with unsupervised data and a reconstruction criterion to extract features from an input image, and the features are used to train a realtime classifier to predict traversability. The online supervision is given by a stereo module that provides robust labels for nearby areas up to 12 meters distant. The approach was developed and tested on the LAGR mobile robot.

AB - We present a learning-based approach for longrange vision that is able to accurately classify complex terrain at distances up to the horizon, thus allowing high-level strategic planning. A deep belief network is trained with unsupervised data and a reconstruction criterion to extract features from an input image, and the features are used to train a realtime classifier to predict traversability. The online supervision is given by a stereo module that provides robust labels for nearby areas up to 12 meters distant. The approach was developed and tested on the LAGR mobile robot.

UR - http://www.scopus.com/inward/record.url?scp=69549124128&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=69549124128&partnerID=8YFLogxK

U2 - 10.1109/IROS.2008.4651217

DO - 10.1109/IROS.2008.4651217

M3 - Conference contribution

AN - SCOPUS:69549124128

SN - 9781424420582

SP - 628

EP - 633

BT - 2008 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS

ER -