Publisher:Pasadena, CA : Jet Propulsion Laboratory, National Aeronautics and Space Administration, 2013
Citation:23rd AAS/AIAA Spaceflight Mechanics Meeting, Kauai, Hawaii, February 10-14 2013
Abstract:
We propose using cluster analysis to perform quick screening for combinatorial global optimization problems. The key missing component currently preventing cluster analysis from use in this context is the lack of a useable metric function that defines the cost to transfer between two orbits. We study several proposed metrics and clustering algorithms, including k-means and the expectation maximization algorithm. We also show that proven heuristic methods such as the Q-law can be modified to work with cluster analysis.