Collective behavior evolution in a group of cooperating agents

  • Authors:
  • J. Liu;J. Wu

  • Affiliations:
  • Department of Computer Science, Hong Kong Baptist University, Kowloon Tong, Hong Kong;Intelligent Systems Group, University of Calgary, Alberta, Canada

  • Venue:
  • Intelligent agents and their applications
  • Year:
  • 2002

Quantified Score

Hi-index 0.00

Visualization

Abstract

This work addresses the issue of how to acquire a collective goal-directed task-driven behavior in a group of distributed agents. The specific problem that we consider here is how simulated ants can perform coordinated movements to collectively transport an object toward a desired goal location. We propose an evolutionary computation mechanism in which no centralized modeling and control are involved except a high-level criterion for measuring the quality of collective task performance. The evolutionary learning approach is based on a fittest-preserved genetic algorithm. While giving the formulation and algorithm of collective behavior learning, we also present the results of several computer simulations for illustrating and validating the effectiveness of the proposed approach.