Acquisition of a destination path accompanied by obstacle avoiding actions for quadruped robots using neural network

Tomohiro Yamaguchi; Keigo Watanabe; Kiyotaka Izumi; Kazuo Kiguchi

doi:10.1299/kikaic.69.2880

Acquisition of a destination path accompanied by obstacle avoiding actions for quadruped robots using neural network

Tomohiro Yamaguchi, Keigo Watanabe, Kiyotaka Izumi, Kazuo Kiguchi

Research output: Contribution to journal › Article › peer-review

Abstract

In the obstacle avoidance of a legged type robot, it is not necessary to avoid all of obstacles by only turning, because it can get over or stride some of them, depending on the obstacle configuration and the state of the robot, unlike a wheel type or a crawler type robot. It is thought that the mobility efficiency to the destination is improved by getting-over or striding. In this paper, a neural network (NN) is used to determine the action of a quadruped robot in the obstacle avoidance path by using information on the destination, the obstacle configuration, and the robot's self-state. The design parameters of NN are adjusted by genetic algorithm (GA) offline. The effectiveness of the present method is proved through an actual experiment.

Original language	English
Pages (from-to)	2880-2887
Number of pages	8
Journal	Nippon Kikai Gakkai Ronbunshu, C Hen/Transactions of the Japan Society of Mechanical Engineers, Part C
Volume	69
Issue number	11
DOIs	https://doi.org/10.1299/kikaic.69.2880
Publication status	Published - Nov 2003
Externally published	Yes

Keywords

Genetic algorithm
Neural network
Obstacle avoidance
Quadruped robot

ASJC Scopus subject areas

Mechanics of Materials
Mechanical Engineering
Industrial and Manufacturing Engineering

Access to Document

10.1299/kikaic.69.2880

Cite this

Acquisition of a destination path accompanied by obstacle avoiding actions for quadruped robots using neural network. / Yamaguchi, Tomohiro; Watanabe, Keigo; Izumi, Kiyotaka et al.
In: Nippon Kikai Gakkai Ronbunshu, C Hen/Transactions of the Japan Society of Mechanical Engineers, Part C, Vol. 69, No. 11, 11.2003, p. 2880-2887.

Research output: Contribution to journal › Article › peer-review

@article{e4b789802aca4fa481fca1d1bfb24479,

title = "Acquisition of a destination path accompanied by obstacle avoiding actions for quadruped robots using neural network",

abstract = "In the obstacle avoidance of a legged type robot, it is not necessary to avoid all of obstacles by only turning, because it can get over or stride some of them, depending on the obstacle configuration and the state of the robot, unlike a wheel type or a crawler type robot. It is thought that the mobility efficiency to the destination is improved by getting-over or striding. In this paper, a neural network (NN) is used to determine the action of a quadruped robot in the obstacle avoidance path by using information on the destination, the obstacle configuration, and the robot's self-state. The design parameters of NN are adjusted by genetic algorithm (GA) offline. The effectiveness of the present method is proved through an actual experiment.",

keywords = "Genetic algorithm, Neural network, Obstacle avoidance, Quadruped robot",

author = "Tomohiro Yamaguchi and Keigo Watanabe and Kiyotaka Izumi and Kazuo Kiguchi",

year = "2003",

month = nov,

doi = "10.1299/kikaic.69.2880",

language = "English",

volume = "69",

pages = "2880--2887",

journal = "Nippon Kikai Gakkai Ronbunshu, C Hen/Transactions of the Japan Society of Mechanical Engineers, Part C",

issn = "0387-5024",

publisher = "Japan Society of Mechanical Engineers",

number = "11",

}

TY - JOUR

T1 - Acquisition of a destination path accompanied by obstacle avoiding actions for quadruped robots using neural network

AU - Yamaguchi, Tomohiro

AU - Watanabe, Keigo

AU - Izumi, Kiyotaka

AU - Kiguchi, Kazuo

PY - 2003/11

Y1 - 2003/11

N2 - In the obstacle avoidance of a legged type robot, it is not necessary to avoid all of obstacles by only turning, because it can get over or stride some of them, depending on the obstacle configuration and the state of the robot, unlike a wheel type or a crawler type robot. It is thought that the mobility efficiency to the destination is improved by getting-over or striding. In this paper, a neural network (NN) is used to determine the action of a quadruped robot in the obstacle avoidance path by using information on the destination, the obstacle configuration, and the robot's self-state. The design parameters of NN are adjusted by genetic algorithm (GA) offline. The effectiveness of the present method is proved through an actual experiment.

AB - In the obstacle avoidance of a legged type robot, it is not necessary to avoid all of obstacles by only turning, because it can get over or stride some of them, depending on the obstacle configuration and the state of the robot, unlike a wheel type or a crawler type robot. It is thought that the mobility efficiency to the destination is improved by getting-over or striding. In this paper, a neural network (NN) is used to determine the action of a quadruped robot in the obstacle avoidance path by using information on the destination, the obstacle configuration, and the robot's self-state. The design parameters of NN are adjusted by genetic algorithm (GA) offline. The effectiveness of the present method is proved through an actual experiment.

KW - Genetic algorithm

KW - Neural network

KW - Obstacle avoidance

KW - Quadruped robot

UR - http://www.scopus.com/inward/record.url?scp=0742321244&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0742321244&partnerID=8YFLogxK

U2 - 10.1299/kikaic.69.2880

DO - 10.1299/kikaic.69.2880

M3 - Article

AN - SCOPUS:0742321244

SN - 0387-5024

VL - 69

SP - 2880

EP - 2887

JO - Nippon Kikai Gakkai Ronbunshu, C Hen/Transactions of the Japan Society of Mechanical Engineers, Part C

JF - Nippon Kikai Gakkai Ronbunshu, C Hen/Transactions of the Japan Society of Mechanical Engineers, Part C

IS - 11

ER -

Acquisition of a destination path accompanied by obstacle avoiding actions for quadruped robots using neural network

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this