Motion imitation learning and real-time movement generation of humanoid using evolutionary algorithm

Ga lam Park, Syung kwon Ra, Chang hwan Kim, Jae bok Song

Research output: Contribution to journalArticle

Abstract

This paper presents a framework to generate human-like movements of a humanoid in real time using the movement primitive database of a human. The framework consists of two processes: 1) the offline motion imitation learning based on an Evolutionary Algorithm and 2) the online motion generation of a humanoid using the database updated by the motion imitation learning. For the offline process, the initial database contains the kinetic characteristics of a human, since it is full of human's captured motions. The database then develops through the proposed framework of motion learning based on an Evolutionary Algorithm, having the kinetic characteristics of a humanoid in aspect of minimal torque or joint jerk. The humanoid generates human-like movements for a given purpose in real time by linearly interpolating the primitive motions in the developed database. The movement of catching a ball was examined in simulation.

Original languageEnglish
Pages (from-to)1038-1046
Number of pages9
JournalJournal of Institute of Control, Robotics and Systems
Volume14
Issue number10
DOIs
Publication statusPublished - 2008 Oct 1

Fingerprint

Imitation
Evolutionary algorithms
Evolutionary Algorithms
Real-time
Motion
Kinetics
Torque
Learning
Movement
Ball
Linearly
Human
Framework
Simulation

Keywords

  • Evolutionary algorithm
  • Human-like movement
  • Humanoid
  • Imitation learning

ASJC Scopus subject areas

  • Software
  • Control and Systems Engineering
  • Applied Mathematics

Cite this

Motion imitation learning and real-time movement generation of humanoid using evolutionary algorithm. / Park, Ga lam; Ra, Syung kwon; Kim, Chang hwan; Song, Jae bok.

In: Journal of Institute of Control, Robotics and Systems, Vol. 14, No. 10, 01.10.2008, p. 1038-1046.

Research output: Contribution to journalArticle

@article{7a367cefca6a414386eec8f67b64f06c,
title = "Motion imitation learning and real-time movement generation of humanoid using evolutionary algorithm",
abstract = "This paper presents a framework to generate human-like movements of a humanoid in real time using the movement primitive database of a human. The framework consists of two processes: 1) the offline motion imitation learning based on an Evolutionary Algorithm and 2) the online motion generation of a humanoid using the database updated by the motion imitation learning. For the offline process, the initial database contains the kinetic characteristics of a human, since it is full of human's captured motions. The database then develops through the proposed framework of motion learning based on an Evolutionary Algorithm, having the kinetic characteristics of a humanoid in aspect of minimal torque or joint jerk. The humanoid generates human-like movements for a given purpose in real time by linearly interpolating the primitive motions in the developed database. The movement of catching a ball was examined in simulation.",
keywords = "Evolutionary algorithm, Human-like movement, Humanoid, Imitation learning",
author = "Park, {Ga lam} and Ra, {Syung kwon} and Kim, {Chang hwan} and Song, {Jae bok}",
year = "2008",
month = "10",
day = "1",
doi = "10.5302/J.ICROS.2008.14.10.1038",
language = "English",
volume = "14",
pages = "1038--1046",
journal = "Journal of Institute of Control, Robotics and Systems",
issn = "1976-5622",
publisher = "Institute of Control, Robotics and Systems",
number = "10",

}

TY - JOUR

T1 - Motion imitation learning and real-time movement generation of humanoid using evolutionary algorithm

AU - Park, Ga lam

AU - Ra, Syung kwon

AU - Kim, Chang hwan

AU - Song, Jae bok

PY - 2008/10/1

Y1 - 2008/10/1

N2 - This paper presents a framework to generate human-like movements of a humanoid in real time using the movement primitive database of a human. The framework consists of two processes: 1) the offline motion imitation learning based on an Evolutionary Algorithm and 2) the online motion generation of a humanoid using the database updated by the motion imitation learning. For the offline process, the initial database contains the kinetic characteristics of a human, since it is full of human's captured motions. The database then develops through the proposed framework of motion learning based on an Evolutionary Algorithm, having the kinetic characteristics of a humanoid in aspect of minimal torque or joint jerk. The humanoid generates human-like movements for a given purpose in real time by linearly interpolating the primitive motions in the developed database. The movement of catching a ball was examined in simulation.

AB - This paper presents a framework to generate human-like movements of a humanoid in real time using the movement primitive database of a human. The framework consists of two processes: 1) the offline motion imitation learning based on an Evolutionary Algorithm and 2) the online motion generation of a humanoid using the database updated by the motion imitation learning. For the offline process, the initial database contains the kinetic characteristics of a human, since it is full of human's captured motions. The database then develops through the proposed framework of motion learning based on an Evolutionary Algorithm, having the kinetic characteristics of a humanoid in aspect of minimal torque or joint jerk. The humanoid generates human-like movements for a given purpose in real time by linearly interpolating the primitive motions in the developed database. The movement of catching a ball was examined in simulation.

KW - Evolutionary algorithm

KW - Human-like movement

KW - Humanoid

KW - Imitation learning

UR - http://www.scopus.com/inward/record.url?scp=84860431102&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84860431102&partnerID=8YFLogxK

U2 - 10.5302/J.ICROS.2008.14.10.1038

DO - 10.5302/J.ICROS.2008.14.10.1038

M3 - Article

AN - SCOPUS:84860431102

VL - 14

SP - 1038

EP - 1046

JO - Journal of Institute of Control, Robotics and Systems

JF - Journal of Institute of Control, Robotics and Systems

SN - 1976-5622

IS - 10

ER -