A cooperative motion control of 2-dof robot arms by neuro-evolved agents

Yingda Dai, Masami Konishi, Jun Imai

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

2 Citations (Scopus)

Abstract

In the paper we propose the method of agent-based neural network model to enhance the behaviors and the communication functions of real planner two degrees of freedom (2-dof) robot arms. Each joint of the manipulator is respectively provided a learning method to optimize trajectory by training RNN model. The evolutionary process in our experiments is carried out entirely on the robot by the proposed controller without human intervention. In addition, the master/slave manipulator system is proposed. The slave arm cooperates with the master like the action of human. Each joint is controlled by the distributed subagent. The method is first evaluated on a relatively simple task and then on increasingly complex behaviors towards the goal tasks. Simulation results show the effectiveness of this approach, and that the proposed RNN model can successfully learning the inverse dynamics of robot manipulators, perform accurate tracking for a general trajectory.

Original languageEnglish
Title of host publicationSecond International Conference on Innovative Computing, Information and Control, ICICIC 2007
DOIs
Publication statusPublished - 2008
Event2nd International Conference on Innovative Computing, Information and Control, ICICIC 2007 - Kumamoto, Japan
Duration: Sep 5 2007Sep 7 2007

Other

Other2nd International Conference on Innovative Computing, Information and Control, ICICIC 2007
CountryJapan
CityKumamoto
Period9/5/079/7/07

Fingerprint

Motion control
Manipulators
Robots
Trajectories
Neural networks
Controllers
Communication
Experiments

ASJC Scopus subject areas

  • Computer Science(all)
  • Mechanical Engineering

Cite this

Dai, Y., Konishi, M., & Imai, J. (2008). A cooperative motion control of 2-dof robot arms by neuro-evolved agents. In Second International Conference on Innovative Computing, Information and Control, ICICIC 2007 [4427754] https://doi.org/10.1109/ICICIC.2007.11

A cooperative motion control of 2-dof robot arms by neuro-evolved agents. / Dai, Yingda; Konishi, Masami; Imai, Jun.

Second International Conference on Innovative Computing, Information and Control, ICICIC 2007. 2008. 4427754.

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

Dai, Y, Konishi, M & Imai, J 2008, A cooperative motion control of 2-dof robot arms by neuro-evolved agents. in Second International Conference on Innovative Computing, Information and Control, ICICIC 2007., 4427754, 2nd International Conference on Innovative Computing, Information and Control, ICICIC 2007, Kumamoto, Japan, 9/5/07. https://doi.org/10.1109/ICICIC.2007.11
Dai Y, Konishi M, Imai J. A cooperative motion control of 2-dof robot arms by neuro-evolved agents. In Second International Conference on Innovative Computing, Information and Control, ICICIC 2007. 2008. 4427754 https://doi.org/10.1109/ICICIC.2007.11
Dai, Yingda ; Konishi, Masami ; Imai, Jun. / A cooperative motion control of 2-dof robot arms by neuro-evolved agents. Second International Conference on Innovative Computing, Information and Control, ICICIC 2007. 2008.
@inproceedings{0737c1188a4342e592df54a11a0be84e,
title = "A cooperative motion control of 2-dof robot arms by neuro-evolved agents",
abstract = "In the paper we propose the method of agent-based neural network model to enhance the behaviors and the communication functions of real planner two degrees of freedom (2-dof) robot arms. Each joint of the manipulator is respectively provided a learning method to optimize trajectory by training RNN model. The evolutionary process in our experiments is carried out entirely on the robot by the proposed controller without human intervention. In addition, the master/slave manipulator system is proposed. The slave arm cooperates with the master like the action of human. Each joint is controlled by the distributed subagent. The method is first evaluated on a relatively simple task and then on increasingly complex behaviors towards the goal tasks. Simulation results show the effectiveness of this approach, and that the proposed RNN model can successfully learning the inverse dynamics of robot manipulators, perform accurate tracking for a general trajectory.",
author = "Yingda Dai and Masami Konishi and Jun Imai",
year = "2008",
doi = "10.1109/ICICIC.2007.11",
language = "English",
isbn = "0769528821",
booktitle = "Second International Conference on Innovative Computing, Information and Control, ICICIC 2007",

}

TY - GEN

T1 - A cooperative motion control of 2-dof robot arms by neuro-evolved agents

AU - Dai, Yingda

AU - Konishi, Masami

AU - Imai, Jun

PY - 2008

Y1 - 2008

N2 - In the paper we propose the method of agent-based neural network model to enhance the behaviors and the communication functions of real planner two degrees of freedom (2-dof) robot arms. Each joint of the manipulator is respectively provided a learning method to optimize trajectory by training RNN model. The evolutionary process in our experiments is carried out entirely on the robot by the proposed controller without human intervention. In addition, the master/slave manipulator system is proposed. The slave arm cooperates with the master like the action of human. Each joint is controlled by the distributed subagent. The method is first evaluated on a relatively simple task and then on increasingly complex behaviors towards the goal tasks. Simulation results show the effectiveness of this approach, and that the proposed RNN model can successfully learning the inverse dynamics of robot manipulators, perform accurate tracking for a general trajectory.

AB - In the paper we propose the method of agent-based neural network model to enhance the behaviors and the communication functions of real planner two degrees of freedom (2-dof) robot arms. Each joint of the manipulator is respectively provided a learning method to optimize trajectory by training RNN model. The evolutionary process in our experiments is carried out entirely on the robot by the proposed controller without human intervention. In addition, the master/slave manipulator system is proposed. The slave arm cooperates with the master like the action of human. Each joint is controlled by the distributed subagent. The method is first evaluated on a relatively simple task and then on increasingly complex behaviors towards the goal tasks. Simulation results show the effectiveness of this approach, and that the proposed RNN model can successfully learning the inverse dynamics of robot manipulators, perform accurate tracking for a general trajectory.

UR - http://www.scopus.com/inward/record.url?scp=39049108357&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=39049108357&partnerID=8YFLogxK

U2 - 10.1109/ICICIC.2007.11

DO - 10.1109/ICICIC.2007.11

M3 - Conference contribution

AN - SCOPUS:39049108357

SN - 0769528821

SN - 9780769528823

BT - Second International Conference on Innovative Computing, Information and Control, ICICIC 2007

ER -