Learning maneuver dictionaries for ground robot planning

Pierre Sermanet, Marco Scoffier, Chris Crudele, Urs Muller, Yann LeCun

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

Abstract

Vehicle dynamics is typically handled by models whose parameters are found through system identification or manually computed from the vehicle's characteristics. While these methods provide accurate theoretical dynamical models, they may not take into account differences between individual vehicles, lack adaptability to new environments and may not handle sophisticated models, requiring hand-crafted heuristics for backwards motion for example. Similarly to space and aerial maneuver-based planning methods, we demonstrate a simple and computationally fast planning method for ground robots with obstacle avoidance. It bypasses the need for model parameters identification and hand-crafted heuristics, learns the particularities of individual vehicles, allows on-line adaptation and sophisticated models. Human-driven or autonomouslydriven trajectories are recorded and stored into a trajectory bank. While in learning mode, the robot records each traveled trajectory and places it into a bank, indexed by the initial speeds of each left and right wheels and the ending position at a fixed radius. Only the best trajectories are stored in the trajectory bank and then reused during autonomous runs for optimal short-range planning. Pre-computed (but not recorded) trajectories have been used in previous work and provide an important computational advantage over on-line computation methods, which are less practical in real-time applications due to the high-dimensional search space. A collision-free platform was developed without any hand-crafted heuristics or knowledge about the vehicle's characteristics. This method is demonstrated on the LAGR platform, a non-holonomic (differential drive) off-road mobile robot.

Original languageEnglish (US)
Title of host publication39th International Symposium on Robotics, ISR 2008
Pages962-967
Number of pages6
StatePublished - 2008
Event39th International Symposium on Robotics, ISR 2008 - Seoul, Korea, Republic of
Duration: Oct 15 2008Oct 17 2008

Other

Other39th International Symposium on Robotics, ISR 2008
CountryKorea, Republic of
CitySeoul
Period10/15/0810/17/08

Fingerprint

Glossaries
Trajectories
Robots
Planning
Identification (control systems)
Collision avoidance
Mobile robots
Wheels
Antennas

ASJC Scopus subject areas

  • Artificial Intelligence
  • Human-Computer Interaction
  • Software

Cite this

Sermanet, P., Scoffier, M., Crudele, C., Muller, U., & LeCun, Y. (2008). Learning maneuver dictionaries for ground robot planning. In 39th International Symposium on Robotics, ISR 2008 (pp. 962-967)

Learning maneuver dictionaries for ground robot planning. / Sermanet, Pierre; Scoffier, Marco; Crudele, Chris; Muller, Urs; LeCun, Yann.

39th International Symposium on Robotics, ISR 2008. 2008. p. 962-967.

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

Sermanet, P, Scoffier, M, Crudele, C, Muller, U & LeCun, Y 2008, Learning maneuver dictionaries for ground robot planning. in 39th International Symposium on Robotics, ISR 2008. pp. 962-967, 39th International Symposium on Robotics, ISR 2008, Seoul, Korea, Republic of, 10/15/08.
Sermanet P, Scoffier M, Crudele C, Muller U, LeCun Y. Learning maneuver dictionaries for ground robot planning. In 39th International Symposium on Robotics, ISR 2008. 2008. p. 962-967
Sermanet, Pierre ; Scoffier, Marco ; Crudele, Chris ; Muller, Urs ; LeCun, Yann. / Learning maneuver dictionaries for ground robot planning. 39th International Symposium on Robotics, ISR 2008. 2008. pp. 962-967
@inproceedings{6dabb540ade54f11a392932f407d802a,
title = "Learning maneuver dictionaries for ground robot planning",
abstract = "Vehicle dynamics is typically handled by models whose parameters are found through system identification or manually computed from the vehicle's characteristics. While these methods provide accurate theoretical dynamical models, they may not take into account differences between individual vehicles, lack adaptability to new environments and may not handle sophisticated models, requiring hand-crafted heuristics for backwards motion for example. Similarly to space and aerial maneuver-based planning methods, we demonstrate a simple and computationally fast planning method for ground robots with obstacle avoidance. It bypasses the need for model parameters identification and hand-crafted heuristics, learns the particularities of individual vehicles, allows on-line adaptation and sophisticated models. Human-driven or autonomouslydriven trajectories are recorded and stored into a trajectory bank. While in learning mode, the robot records each traveled trajectory and places it into a bank, indexed by the initial speeds of each left and right wheels and the ending position at a fixed radius. Only the best trajectories are stored in the trajectory bank and then reused during autonomous runs for optimal short-range planning. Pre-computed (but not recorded) trajectories have been used in previous work and provide an important computational advantage over on-line computation methods, which are less practical in real-time applications due to the high-dimensional search space. A collision-free platform was developed without any hand-crafted heuristics or knowledge about the vehicle's characteristics. This method is demonstrated on the LAGR platform, a non-holonomic (differential drive) off-road mobile robot.",
author = "Pierre Sermanet and Marco Scoffier and Chris Crudele and Urs Muller and Yann LeCun",
year = "2008",
language = "English (US)",
pages = "962--967",
booktitle = "39th International Symposium on Robotics, ISR 2008",

}

TY - GEN

T1 - Learning maneuver dictionaries for ground robot planning

AU - Sermanet, Pierre

AU - Scoffier, Marco

AU - Crudele, Chris

AU - Muller, Urs

AU - LeCun, Yann

PY - 2008

Y1 - 2008

N2 - Vehicle dynamics is typically handled by models whose parameters are found through system identification or manually computed from the vehicle's characteristics. While these methods provide accurate theoretical dynamical models, they may not take into account differences between individual vehicles, lack adaptability to new environments and may not handle sophisticated models, requiring hand-crafted heuristics for backwards motion for example. Similarly to space and aerial maneuver-based planning methods, we demonstrate a simple and computationally fast planning method for ground robots with obstacle avoidance. It bypasses the need for model parameters identification and hand-crafted heuristics, learns the particularities of individual vehicles, allows on-line adaptation and sophisticated models. Human-driven or autonomouslydriven trajectories are recorded and stored into a trajectory bank. While in learning mode, the robot records each traveled trajectory and places it into a bank, indexed by the initial speeds of each left and right wheels and the ending position at a fixed radius. Only the best trajectories are stored in the trajectory bank and then reused during autonomous runs for optimal short-range planning. Pre-computed (but not recorded) trajectories have been used in previous work and provide an important computational advantage over on-line computation methods, which are less practical in real-time applications due to the high-dimensional search space. A collision-free platform was developed without any hand-crafted heuristics or knowledge about the vehicle's characteristics. This method is demonstrated on the LAGR platform, a non-holonomic (differential drive) off-road mobile robot.

AB - Vehicle dynamics is typically handled by models whose parameters are found through system identification or manually computed from the vehicle's characteristics. While these methods provide accurate theoretical dynamical models, they may not take into account differences between individual vehicles, lack adaptability to new environments and may not handle sophisticated models, requiring hand-crafted heuristics for backwards motion for example. Similarly to space and aerial maneuver-based planning methods, we demonstrate a simple and computationally fast planning method for ground robots with obstacle avoidance. It bypasses the need for model parameters identification and hand-crafted heuristics, learns the particularities of individual vehicles, allows on-line adaptation and sophisticated models. Human-driven or autonomouslydriven trajectories are recorded and stored into a trajectory bank. While in learning mode, the robot records each traveled trajectory and places it into a bank, indexed by the initial speeds of each left and right wheels and the ending position at a fixed radius. Only the best trajectories are stored in the trajectory bank and then reused during autonomous runs for optimal short-range planning. Pre-computed (but not recorded) trajectories have been used in previous work and provide an important computational advantage over on-line computation methods, which are less practical in real-time applications due to the high-dimensional search space. A collision-free platform was developed without any hand-crafted heuristics or knowledge about the vehicle's characteristics. This method is demonstrated on the LAGR platform, a non-holonomic (differential drive) off-road mobile robot.

UR - http://www.scopus.com/inward/record.url?scp=84876778347&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84876778347&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84876778347

SP - 962

EP - 967

BT - 39th International Symposium on Robotics, ISR 2008

ER -