Automatic path search for roving robot using reinforcement learning

Shigeharu Miyata, Akira Yanou, Hitomi Nakamura, Shin Takehara

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

Rapid advances in robot technology have been made in recent years. Inconnection with these advances, robots are expected, to be utilized, in a,variety of places and, environments. This study describes 1) a, method whichallows a, robot to measure the location of its destination in the real world,based, on an image obtained, from, a, single camera, and, 2) a, method ofnavigating a, robot to a, destination which is selected, by a, user on a,display showing the forward, robot view. Consideration is also given to cases inwhich there are obstacles between the robot and, the destination. Through theuse of reinforcement learning, which is considered, a, promising candidate amongautonomous control techniques, the roving robot tries to find the shortest wayto the destination based on information concerning the locations of obstaclesand, the destination. This study also describes an image-based method ofmeasuring a, selected, location, the results from, a simulation of path findingusing reinforcement learning, and the results from an experiment of navigationin a, real environment. Finally, a, summary of the main conclusions is provided.ICIC International

Original languageEnglish
Pages (from-to)885-892
Number of pages8
JournalICIC Express Letters
Volume4
Issue number3
Publication statusPublished - Jun 2010

Fingerprint

Reinforcement learning
Robots
Cameras
Display devices

Keywords

  • Navigation
  • Path finding
  • Reinforcement learning
  • Roving robot

ASJC Scopus subject areas

  • Computer Science(all)
  • Control and Systems Engineering

Cite this

Miyata, S., Yanou, A., Nakamura, H., & Takehara, S. (2010). Automatic path search for roving robot using reinforcement learning. ICIC Express Letters, 4(3), 885-892.

Automatic path search for roving robot using reinforcement learning. / Miyata, Shigeharu; Yanou, Akira; Nakamura, Hitomi; Takehara, Shin.

In: ICIC Express Letters, Vol. 4, No. 3, 06.2010, p. 885-892.

Research output: Contribution to journalArticle

Miyata, S, Yanou, A, Nakamura, H & Takehara, S 2010, 'Automatic path search for roving robot using reinforcement learning', ICIC Express Letters, vol. 4, no. 3, pp. 885-892.
Miyata S, Yanou A, Nakamura H, Takehara S. Automatic path search for roving robot using reinforcement learning. ICIC Express Letters. 2010 Jun;4(3):885-892.
Miyata, Shigeharu ; Yanou, Akira ; Nakamura, Hitomi ; Takehara, Shin. / Automatic path search for roving robot using reinforcement learning. In: ICIC Express Letters. 2010 ; Vol. 4, No. 3. pp. 885-892.
@article{ce1bcb4151694ccba157fc9800cf6c35,
title = "Automatic path search for roving robot using reinforcement learning",
abstract = "Rapid advances in robot technology have been made in recent years. Inconnection with these advances, robots are expected, to be utilized, in a,variety of places and, environments. This study describes 1) a, method whichallows a, robot to measure the location of its destination in the real world,based, on an image obtained, from, a, single camera, and, 2) a, method ofnavigating a, robot to a, destination which is selected, by a, user on a,display showing the forward, robot view. Consideration is also given to cases inwhich there are obstacles between the robot and, the destination. Through theuse of reinforcement learning, which is considered, a, promising candidate amongautonomous control techniques, the roving robot tries to find the shortest wayto the destination based on information concerning the locations of obstaclesand, the destination. This study also describes an image-based method ofmeasuring a, selected, location, the results from, a simulation of path findingusing reinforcement learning, and the results from an experiment of navigationin a, real environment. Finally, a, summary of the main conclusions is provided.ICIC International",
keywords = "Navigation, Path finding, Reinforcement learning, Roving robot",
author = "Shigeharu Miyata and Akira Yanou and Hitomi Nakamura and Shin Takehara",
year = "2010",
month = "6",
language = "English",
volume = "4",
pages = "885--892",
journal = "ICIC Express Letters",
issn = "1881-803X",
publisher = "ICIC Express Letters Office",
number = "3",

}

TY - JOUR

T1 - Automatic path search for roving robot using reinforcement learning

AU - Miyata, Shigeharu

AU - Yanou, Akira

AU - Nakamura, Hitomi

AU - Takehara, Shin

PY - 2010/6

Y1 - 2010/6

N2 - Rapid advances in robot technology have been made in recent years. Inconnection with these advances, robots are expected, to be utilized, in a,variety of places and, environments. This study describes 1) a, method whichallows a, robot to measure the location of its destination in the real world,based, on an image obtained, from, a, single camera, and, 2) a, method ofnavigating a, robot to a, destination which is selected, by a, user on a,display showing the forward, robot view. Consideration is also given to cases inwhich there are obstacles between the robot and, the destination. Through theuse of reinforcement learning, which is considered, a, promising candidate amongautonomous control techniques, the roving robot tries to find the shortest wayto the destination based on information concerning the locations of obstaclesand, the destination. This study also describes an image-based method ofmeasuring a, selected, location, the results from, a simulation of path findingusing reinforcement learning, and the results from an experiment of navigationin a, real environment. Finally, a, summary of the main conclusions is provided.ICIC International

AB - Rapid advances in robot technology have been made in recent years. Inconnection with these advances, robots are expected, to be utilized, in a,variety of places and, environments. This study describes 1) a, method whichallows a, robot to measure the location of its destination in the real world,based, on an image obtained, from, a, single camera, and, 2) a, method ofnavigating a, robot to a, destination which is selected, by a, user on a,display showing the forward, robot view. Consideration is also given to cases inwhich there are obstacles between the robot and, the destination. Through theuse of reinforcement learning, which is considered, a, promising candidate amongautonomous control techniques, the roving robot tries to find the shortest wayto the destination based on information concerning the locations of obstaclesand, the destination. This study also describes an image-based method ofmeasuring a, selected, location, the results from, a simulation of path findingusing reinforcement learning, and the results from an experiment of navigationin a, real environment. Finally, a, summary of the main conclusions is provided.ICIC International

KW - Navigation

KW - Path finding

KW - Reinforcement learning

KW - Roving robot

UR - http://www.scopus.com/inward/record.url?scp=77953902842&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77953902842&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:77953902842

VL - 4

SP - 885

EP - 892

JO - ICIC Express Letters

JF - ICIC Express Letters

SN - 1881-803X

IS - 3

ER -