Pursuit Domain
The example runs below show how RMM agents, using modeling down to the fourth level of nesting, coordinate with each other, and with humans, to surround and capture a "prey". We have included the demonstration of the mixed RMM/human coordination to show the viability of modeling human action as rational, in spite of humans simply following their intuitions in deciding on actions, and not attempting to figure out what the best, rational action is.
Four RMM agents trying to capture a prey. The agents are numbered 1 through 4. P is the prey.
Three RMM agents, coordinating with one human-controlled agent, are capturing a prey. RMM agents are numbered 1 through 3.The human agent has the number 4. P is the prey.
Two RMM agents and two human agents capturing the prey P. Number 1 to 2 are the RMM agents. 3 and 4 are the human agents.
A single RMM agent (number 1) is coordinating with three human agents (numbered 2 through 4). For comparison, this is how four human agents coordinate (with no communication) in capturing the prey P.MISSION · PUBLICATION · MEMBER · RESEARCH · LINKS
© 1998 Multi-Agent Systems Group at UIC