Improving Aircraft Collision Risk Estimation Using the Cross-Entropy Method

Estimates of mid-air collision risk may be obtained through Monte Carlo simulation of encounters sampled from a probabilistic airspace encounter model. Due to the rarity of collision events, typically millions of simulations are required. A technique known as importance sampling has been used in the past to bias the sampling on trajectories that are likely to result in collisions. However, determining a suitable importance sampling distribution is not straightforward. This paper explores the use of the cross-entropy method to adaptively determine a suitable sampling distribution and analyzes its performance. The results show that reliable estimates of collision risk can be obtained with only a fraction of the computational cost required by methods used in past studies.

  • Record URL:
  • Supplemental Notes:
    • Copyright © 2016 by Stanford University. Published by the American Institute of Aeronautics and Astronautics, Inc., with permission. Copies of this paper may be made for personal and internal use, on condition that the copier pay the per-copy fee to the Copyright Clearance Center (CCC). All requests for copying and permission to reprint should be submitted to CCC at; employ the ISSN 2380-9450 (online) to initiate your request.
  • Authors:
    • Kim, Youngjun
    • Kochenderfer, Mykel J
  • Publication Date: 2016-4


  • English

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 01768974
  • Record Type: Publication
  • Files: TRIS
  • Created Date: Mar 9 2021 6:49AM