An Evolutionary Solution for Cooperative and Competitive Mobile Agents

  • Authors:

  • Jiancong Fan;Jiuhong Ruan;Yongquan Liang

  • Affiliations:

  • College of Information Science and Engineering, Shandong University of Science and Technology, Qingdao, China 266510;Scientific Research Department, Shandong Jiaotong University, Jinan, China 250023;College of Information Science and Engineering, Shandong University of Science and Technology, Qingdao, China 266510

  • Venue:
  • AICI '09 Proceedings of the International Conference on Artificial Intelligence and Computational Intelligence
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

The cooperation and competition among mobile agents using evolutionary strategy is an important domain in Agent theory and application. With evolutionary strategy the cooperation process is achieved by training and iterating many times. From evolutionary solution of cooperative and competitive mobile agents (CCMA), a group of mobile agents are partitioned into two populations, cooperative agents group and competitive agent group. Cooperative agents are treated as several pursuers, while a competitive agent is viewed as the pursuers' competitor called evader. The cooperation actions take place among the pursuers in order to capture the evader as rapidly as possible. An agent individual (chromosome) is encoded based on a kind of two-dimensional random moving. The next moving direction is encoded as chromosome. The chromosome can be crossed over and mutated according to designed operators and fitness function. An evolutionary algorithm for cooperation and competition of mobile agents is proposed. The experiments show that the algorithm for this evolutionary solution is effective, and it has better time performances and convergence.