Research Papers

Search Strategies in Evolutionary Multi-Agent Systems: The Effect of Cooperation and Reward on Solution Quality

[+] Author and Article Information
Lindsay Hanna Landry

Jonathan Cagan1

Department of Mechanical Engineering,  Carnegie Mellon University, Pittsburgh, PA 15213jcag@andrew.cmu.edu

In the original version of EMAS [12], there were multiple shared memories and an agent's chromosome identified which memory it would use. With multiple memories, cooperation is not guaranteed between all agents, so to isolate and examine the effect of cooperation a single memory is used here.


Corresponding author.

J. Mech. Des 133(6), 061005 (Jun 15, 2011) (8 pages) doi:10.1115/1.4004192 History: Received October 19, 2009; Revised April 20, 2011; Published June 15, 2011; Online June 15, 2011

Cooperation and reward of strategic agents in an evolutionary optimization framework is explored in order to better solve engineering design problems. Agents in this Evolutionary Multi-Agent Systems (EMAS) framework rely on one another to better their performance, but also vie for the opportunity to reproduce. The level of cooperation and reward is varied by altering the amount of interaction between agents and the fitness function describing their evolution. The effect of each variable is measured using the problem objective function as a metric. Increasing the amount of cooperation in the evolving team is shown to lead to improved performance for several multimodal and complex numerical optimization and three-dimensional layout problems. However, fitness functions that utilize team-based rewards are found to be inferior to those that reward on an individual basis. The performance trends for different fitness functions and levels of cooperation remain when EMAS is applied to the more complex problem of three-dimensional packing as well.

Copyright © 2011 by American Society of Mechanical Engineers
Your Session has timed out. Please sign back in to continue.



Grahic Jump Location
Figure 1

Graphical representation of EMAS implementation for a single iteration. Lines represent the flow of solutions from the shared memory (top) to the agents (middle), from the solution space (bottom) to the agents, and from the agents to the shared memory.

Grahic Jump Location
Figure 2

EMAS agent chromosome representation

Grahic Jump Location
Figure 3

Function spaces used for studying EMAS for numerical optimization: (a) Bohachevsky function, (b) sinusoidal function, (c) peaks function, and (d) fractal function

Grahic Jump Location
Figure 4

Sample packings for the eight-cube problem (left) and for the arbitrary components problem (right). Intersections are shown in red/light gray and protrusions are shown in blue/dark gray.

Grahic Jump Location
Figure 5

Average and standard deviation of percentage of team running each algorithm over the course of 25 iterations of EMAS. Data presented for sinusoidal function under partial cooperation.



Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In