Generation of evaluation function for robot force control using genetic programming

Kazuo Kiguchi; Keigo Watanabe; Kiyotaka Izumi; Toshio Fukuda

Generation of evaluation function for robot force control using genetic programming

Kazuo Kiguchi, Keigo Watanabe, Kiyotaka Izumi, Toshio Fukuda

Research output: Contribution to conference › Paper › peer-review

Abstract

Force control is one of the most important and fundamental tasks of robot manipulators. It is known that a neuro-fuzzy control method is one of the best control methods for robot force control. Usually, the neuro-fuzzy controller is trained to minimize the error function. However, unexpected response such as overshooting or oscillation might occur as far as only the control error is evaluated, since the dynamics of the robot and the environment is not reflected in the evaluation function. In this paper, we propose an effective evaluation function generation method using genetic programming. The effectiveness of the proposed method was evaluated by simulation.

Original language	English
Pages	2767-2771
Number of pages	5
Publication status	Published - Dec 1 2001
Externally published	Yes
Event	Joint 9th IFSA World Congress and 20th NAFIPS International Conference - Vancouver, BC, Canada Duration: Jul 25 2001 → Jul 28 2001

Other

Other	Joint 9th IFSA World Congress and 20th NAFIPS International Conference
Country/Territory	Canada
City	Vancouver, BC
Period	7/25/01 → 7/28/01

All Science Journal Classification (ASJC) codes

Computer Science(all)
Mathematics(all)

Cite this

@conference{799c69d1e1a54aaca9abae7502d36332,

title = "Generation of evaluation function for robot force control using genetic programming",

abstract = "Force control is one of the most important and fundamental tasks of robot manipulators. It is known that a neuro-fuzzy control method is one of the best control methods for robot force control. Usually, the neuro-fuzzy controller is trained to minimize the error function. However, unexpected response such as overshooting or oscillation might occur as far as only the control error is evaluated, since the dynamics of the robot and the environment is not reflected in the evaluation function. In this paper, we propose an effective evaluation function generation method using genetic programming. The effectiveness of the proposed method was evaluated by simulation.",

author = "Kazuo Kiguchi and Keigo Watanabe and Kiyotaka Izumi and Toshio Fukuda",

year = "2001",

month = dec,

day = "1",

language = "English",

pages = "2767--2771",

note = "Joint 9th IFSA World Congress and 20th NAFIPS International Conference ; Conference date: 25-07-2001 Through 28-07-2001",

}

TY - CONF

T1 - Generation of evaluation function for robot force control using genetic programming

AU - Kiguchi, Kazuo

AU - Watanabe, Keigo

AU - Izumi, Kiyotaka

AU - Fukuda, Toshio

PY - 2001/12/1

Y1 - 2001/12/1

N2 - Force control is one of the most important and fundamental tasks of robot manipulators. It is known that a neuro-fuzzy control method is one of the best control methods for robot force control. Usually, the neuro-fuzzy controller is trained to minimize the error function. However, unexpected response such as overshooting or oscillation might occur as far as only the control error is evaluated, since the dynamics of the robot and the environment is not reflected in the evaluation function. In this paper, we propose an effective evaluation function generation method using genetic programming. The effectiveness of the proposed method was evaluated by simulation.

AB - Force control is one of the most important and fundamental tasks of robot manipulators. It is known that a neuro-fuzzy control method is one of the best control methods for robot force control. Usually, the neuro-fuzzy controller is trained to minimize the error function. However, unexpected response such as overshooting or oscillation might occur as far as only the control error is evaluated, since the dynamics of the robot and the environment is not reflected in the evaluation function. In this paper, we propose an effective evaluation function generation method using genetic programming. The effectiveness of the proposed method was evaluated by simulation.

UR - http://www.scopus.com/inward/record.url?scp=0035792825&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0035792825&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:0035792825

SP - 2767

EP - 2771

T2 - Joint 9th IFSA World Congress and 20th NAFIPS International Conference

Y2 - 25 July 2001 through 28 July 2001

ER -

Generation of evaluation function for robot force control using genetic programming

Abstract

Other

All Science Journal Classification (ASJC) codes

Other files and links

Fingerprint

Cite this