Knowledge is a Dangerous Thing: Collective Fluctuations in Pilot Intent

Philip Goetz

It has been shown that, in an abstract model of 2 limited shared resources, when abstract "agents" with similar utility functions for these resources attempt to maximize their private utilities based on obsolete information about what other agents are doing, oscillations or chaotic behavior can arise spontaneously, resulting in lower utility for everyone (Kephart, Hogg, and Huberman 1989). We have seen the same oscillations appear in more realistic simulations of a future air-traffic-management scenario.


This page is copyrighted by AAAI. All rights reserved. Your use of this site constitutes acceptance of all of AAAI's terms and conditions and privacy policy.