Properties of a Formal Method for Prediction of Emergent Behaviors in Swarm-Based Systems

  • Authors:
  • Christopher Rouff;Amy Vanderbilt;Mike Hinchey;Walt Truszkowski;James Rash

  • Affiliations:
  • SAIC;SAIC;NASA GSFC Code 581;NASA GSFC Code 588;NASA GSFC Code 588

  • Venue:
  • SEFM '04 Proceedings of the Software Engineering and Formal Methods, Second International Conference
  • Year:
  • 2004

Quantified Score

Hi-index 0.00

Visualization

Abstract

Autonomous intelligent swarms of satellites are being proposed for NASA missions that have complex behaviors and interactions. The emergent properties of swarms make these missions powerful, but at the same time more difficult to design and assure that proper behaviors will emerge. This paper gives the results of research into formal methods techniques for verification and validation of NASA swarm-based missions. Multiple formal methods were evaluated to determine their effectiveness in modeling and assuring the behavior of swarms of spacecraft. The NASA ANTS mission was used as an example of swarm intelligence for which to apply the formal methods. This paper will give the evaluation of these formal methods and give partial specifications of the ANTS mission using four selected methods. We then give an evaluation of the methods and the needed properties of a formal method for effective specification and prediction of emergent behavior in swarm-based systems.