Coordination of multiple behavior modules evolved on CAM-Brain

K. J. Kim, Sung-Bae Cho

Research output: Contribution to conferencePaper

Abstract

In behavior-based robotics the control of a robot is shared between a set of purposive perception-action units, called behaviors. A major issue in the design of behavior-based control systems is the formulation of effective mechanisms for coordination of the behaviors' activities into strategies for rational and coherent behavior. Early, there has been extensive work to construct an optimal controller for a mobile robot by evolutionary approaches such as genetic algorithm, genetic programming, and so on. In this line of research, we have also presented a method of applying CAM-Brain, evolved neural networks based on cellular automata (CA), to control a mobile robot. However, this approach has a limitation to make the robot to perform appropriate behavior in complex environments. The multi module coordination method can make complex and general behaviors by combining several modules evolved or programmed to do a simple behavior. In this paper, we have attempted to coordinate several modules evolved to do a simple behavior by Maes's Action Selection Mechanism. Maes has proposed a mechanism for action selection, which is reviewed here and is evaulated using a simulated environment. Experimental results show that this approach has potential to develop a sophisticated evolutionary neural controller for complex environments.

Original languageEnglish
Pages1414-1421
Number of pages8
Publication statusPublished - 2001 Jan 1
EventCongress on Evolutionary Computation 2001 - Seoul, Korea, Republic of
Duration: 2001 May 272001 May 30

Other

OtherCongress on Evolutionary Computation 2001
CountryKorea, Republic of
CitySeoul
Period01/5/2701/5/30

Fingerprint

Computer aided manufacturing
Mobile robots
Brain
Robots
Controllers
Genetic programming
Cellular automata
Robotics
Genetic algorithms
Neural networks
Control systems

All Science Journal Classification (ASJC) codes

  • Computer Science(all)
  • Engineering(all)

Cite this

Kim, K. J., & Cho, S-B. (2001). Coordination of multiple behavior modules evolved on CAM-Brain. 1414-1421. Paper presented at Congress on Evolutionary Computation 2001, Seoul, Korea, Republic of.
Kim, K. J. ; Cho, Sung-Bae. / Coordination of multiple behavior modules evolved on CAM-Brain. Paper presented at Congress on Evolutionary Computation 2001, Seoul, Korea, Republic of.8 p.
@conference{e8044ed11ba04cda9204f4a12f98933f,
title = "Coordination of multiple behavior modules evolved on CAM-Brain",
abstract = "In behavior-based robotics the control of a robot is shared between a set of purposive perception-action units, called behaviors. A major issue in the design of behavior-based control systems is the formulation of effective mechanisms for coordination of the behaviors' activities into strategies for rational and coherent behavior. Early, there has been extensive work to construct an optimal controller for a mobile robot by evolutionary approaches such as genetic algorithm, genetic programming, and so on. In this line of research, we have also presented a method of applying CAM-Brain, evolved neural networks based on cellular automata (CA), to control a mobile robot. However, this approach has a limitation to make the robot to perform appropriate behavior in complex environments. The multi module coordination method can make complex and general behaviors by combining several modules evolved or programmed to do a simple behavior. In this paper, we have attempted to coordinate several modules evolved to do a simple behavior by Maes's Action Selection Mechanism. Maes has proposed a mechanism for action selection, which is reviewed here and is evaulated using a simulated environment. Experimental results show that this approach has potential to develop a sophisticated evolutionary neural controller for complex environments.",
author = "Kim, {K. J.} and Sung-Bae Cho",
year = "2001",
month = "1",
day = "1",
language = "English",
pages = "1414--1421",
note = "Congress on Evolutionary Computation 2001 ; Conference date: 27-05-2001 Through 30-05-2001",

}

Kim, KJ & Cho, S-B 2001, 'Coordination of multiple behavior modules evolved on CAM-Brain' Paper presented at Congress on Evolutionary Computation 2001, Seoul, Korea, Republic of, 01/5/27 - 01/5/30, pp. 1414-1421.

Coordination of multiple behavior modules evolved on CAM-Brain. / Kim, K. J.; Cho, Sung-Bae.

2001. 1414-1421 Paper presented at Congress on Evolutionary Computation 2001, Seoul, Korea, Republic of.

Research output: Contribution to conferencePaper

TY - CONF

T1 - Coordination of multiple behavior modules evolved on CAM-Brain

AU - Kim, K. J.

AU - Cho, Sung-Bae

PY - 2001/1/1

Y1 - 2001/1/1

N2 - In behavior-based robotics the control of a robot is shared between a set of purposive perception-action units, called behaviors. A major issue in the design of behavior-based control systems is the formulation of effective mechanisms for coordination of the behaviors' activities into strategies for rational and coherent behavior. Early, there has been extensive work to construct an optimal controller for a mobile robot by evolutionary approaches such as genetic algorithm, genetic programming, and so on. In this line of research, we have also presented a method of applying CAM-Brain, evolved neural networks based on cellular automata (CA), to control a mobile robot. However, this approach has a limitation to make the robot to perform appropriate behavior in complex environments. The multi module coordination method can make complex and general behaviors by combining several modules evolved or programmed to do a simple behavior. In this paper, we have attempted to coordinate several modules evolved to do a simple behavior by Maes's Action Selection Mechanism. Maes has proposed a mechanism for action selection, which is reviewed here and is evaulated using a simulated environment. Experimental results show that this approach has potential to develop a sophisticated evolutionary neural controller for complex environments.

AB - In behavior-based robotics the control of a robot is shared between a set of purposive perception-action units, called behaviors. A major issue in the design of behavior-based control systems is the formulation of effective mechanisms for coordination of the behaviors' activities into strategies for rational and coherent behavior. Early, there has been extensive work to construct an optimal controller for a mobile robot by evolutionary approaches such as genetic algorithm, genetic programming, and so on. In this line of research, we have also presented a method of applying CAM-Brain, evolved neural networks based on cellular automata (CA), to control a mobile robot. However, this approach has a limitation to make the robot to perform appropriate behavior in complex environments. The multi module coordination method can make complex and general behaviors by combining several modules evolved or programmed to do a simple behavior. In this paper, we have attempted to coordinate several modules evolved to do a simple behavior by Maes's Action Selection Mechanism. Maes has proposed a mechanism for action selection, which is reviewed here and is evaulated using a simulated environment. Experimental results show that this approach has potential to develop a sophisticated evolutionary neural controller for complex environments.

UR - http://www.scopus.com/inward/record.url?scp=0034863695&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0034863695&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:0034863695

SP - 1414

EP - 1421

ER -

Kim KJ, Cho S-B. Coordination of multiple behavior modules evolved on CAM-Brain. 2001. Paper presented at Congress on Evolutionary Computation 2001, Seoul, Korea, Republic of.