A multirange architecture for collision-free off-road robot navigation

Pierre Sermanet, Raia Hadsell, Marco Scoffier, Matt Grimes, Jan Ben, Ayse Erkan, Chris Crudele, Urs Miller, Yann LeCun

Research output: Contribution to journalArticle

Abstract

We present a multilayered mapping, planning, and command execution system developed and tested on the LAGR mobile robot. Key to robust performance under uncertainty is the combination of a short-range perception system operating at high frame rate and low resolution and a long-range, adaptive vision system operating at lower frame rate and higher resolution. The short-range module performs local planning and obstacle avoidance with fast reaction times, whereas the long-range module performs strategic visual planning. Probabilistic traversability labels provided by the perception modules are combined and accumulated into a robot-centered hyperbolic-polarmap with a 200-m effective range. Instead of using a dynamical model of the robot for short-range planning, the system uses a large lookup table of physically possible trajectory segments recorded on the robot in a wide variety of driving conditions. Localization is performed using a combination of global positioning system, wheel odometry, inertial measurement unit, and a high-speed, low-complexity rotational visual odometry module. The end-to-end system was developed and tested on the LAGR mobile robot and was verified in independent government tests.

Original languageEnglish (US)
Pages (from-to)52-87
Number of pages36
JournalJournal of Field Robotics
Volume26
Issue number1
DOIs
StatePublished - Jan 2009

Fingerprint

Navigation
Robots
Planning
Mobile robots
Table lookup
Units of measurement
Collision avoidance
Global positioning system
Labels
Wheels
Trajectories

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Computer Science Applications

Cite this

Sermanet, P., Hadsell, R., Scoffier, M., Grimes, M., Ben, J., Erkan, A., ... LeCun, Y. (2009). A multirange architecture for collision-free off-road robot navigation. Journal of Field Robotics, 26(1), 52-87. https://doi.org/10.1002/rob.20270

A multirange architecture for collision-free off-road robot navigation. / Sermanet, Pierre; Hadsell, Raia; Scoffier, Marco; Grimes, Matt; Ben, Jan; Erkan, Ayse; Crudele, Chris; Miller, Urs; LeCun, Yann.

In: Journal of Field Robotics, Vol. 26, No. 1, 01.2009, p. 52-87.

Research output: Contribution to journalArticle

Sermanet, P, Hadsell, R, Scoffier, M, Grimes, M, Ben, J, Erkan, A, Crudele, C, Miller, U & LeCun, Y 2009, 'A multirange architecture for collision-free off-road robot navigation', Journal of Field Robotics, vol. 26, no. 1, pp. 52-87. https://doi.org/10.1002/rob.20270
Sermanet P, Hadsell R, Scoffier M, Grimes M, Ben J, Erkan A et al. A multirange architecture for collision-free off-road robot navigation. Journal of Field Robotics. 2009 Jan;26(1):52-87. https://doi.org/10.1002/rob.20270
Sermanet, Pierre ; Hadsell, Raia ; Scoffier, Marco ; Grimes, Matt ; Ben, Jan ; Erkan, Ayse ; Crudele, Chris ; Miller, Urs ; LeCun, Yann. / A multirange architecture for collision-free off-road robot navigation. In: Journal of Field Robotics. 2009 ; Vol. 26, No. 1. pp. 52-87.
@article{52103140eeb64b18bc5f95538c9ee301,
title = "A multirange architecture for collision-free off-road robot navigation",
abstract = "We present a multilayered mapping, planning, and command execution system developed and tested on the LAGR mobile robot. Key to robust performance under uncertainty is the combination of a short-range perception system operating at high frame rate and low resolution and a long-range, adaptive vision system operating at lower frame rate and higher resolution. The short-range module performs local planning and obstacle avoidance with fast reaction times, whereas the long-range module performs strategic visual planning. Probabilistic traversability labels provided by the perception modules are combined and accumulated into a robot-centered hyperbolic-polarmap with a 200-m effective range. Instead of using a dynamical model of the robot for short-range planning, the system uses a large lookup table of physically possible trajectory segments recorded on the robot in a wide variety of driving conditions. Localization is performed using a combination of global positioning system, wheel odometry, inertial measurement unit, and a high-speed, low-complexity rotational visual odometry module. The end-to-end system was developed and tested on the LAGR mobile robot and was verified in independent government tests.",
author = "Pierre Sermanet and Raia Hadsell and Marco Scoffier and Matt Grimes and Jan Ben and Ayse Erkan and Chris Crudele and Urs Miller and Yann LeCun",
year = "2009",
month = "1",
doi = "10.1002/rob.20270",
language = "English (US)",
volume = "26",
pages = "52--87",
journal = "Journal of Field Robotics",
issn = "1556-4959",
publisher = "John Wiley and Sons Inc.",
number = "1",

}

TY - JOUR

T1 - A multirange architecture for collision-free off-road robot navigation

AU - Sermanet, Pierre

AU - Hadsell, Raia

AU - Scoffier, Marco

AU - Grimes, Matt

AU - Ben, Jan

AU - Erkan, Ayse

AU - Crudele, Chris

AU - Miller, Urs

AU - LeCun, Yann

PY - 2009/1

Y1 - 2009/1

N2 - We present a multilayered mapping, planning, and command execution system developed and tested on the LAGR mobile robot. Key to robust performance under uncertainty is the combination of a short-range perception system operating at high frame rate and low resolution and a long-range, adaptive vision system operating at lower frame rate and higher resolution. The short-range module performs local planning and obstacle avoidance with fast reaction times, whereas the long-range module performs strategic visual planning. Probabilistic traversability labels provided by the perception modules are combined and accumulated into a robot-centered hyperbolic-polarmap with a 200-m effective range. Instead of using a dynamical model of the robot for short-range planning, the system uses a large lookup table of physically possible trajectory segments recorded on the robot in a wide variety of driving conditions. Localization is performed using a combination of global positioning system, wheel odometry, inertial measurement unit, and a high-speed, low-complexity rotational visual odometry module. The end-to-end system was developed and tested on the LAGR mobile robot and was verified in independent government tests.

AB - We present a multilayered mapping, planning, and command execution system developed and tested on the LAGR mobile robot. Key to robust performance under uncertainty is the combination of a short-range perception system operating at high frame rate and low resolution and a long-range, adaptive vision system operating at lower frame rate and higher resolution. The short-range module performs local planning and obstacle avoidance with fast reaction times, whereas the long-range module performs strategic visual planning. Probabilistic traversability labels provided by the perception modules are combined and accumulated into a robot-centered hyperbolic-polarmap with a 200-m effective range. Instead of using a dynamical model of the robot for short-range planning, the system uses a large lookup table of physically possible trajectory segments recorded on the robot in a wide variety of driving conditions. Localization is performed using a combination of global positioning system, wheel odometry, inertial measurement unit, and a high-speed, low-complexity rotational visual odometry module. The end-to-end system was developed and tested on the LAGR mobile robot and was verified in independent government tests.

UR - http://www.scopus.com/inward/record.url?scp=62449226229&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=62449226229&partnerID=8YFLogxK

U2 - 10.1002/rob.20270

DO - 10.1002/rob.20270

M3 - Article

AN - SCOPUS:62449226229

VL - 26

SP - 52

EP - 87

JO - Journal of Field Robotics

JF - Journal of Field Robotics

SN - 1556-4959

IS - 1

ER -