15th Triennial World Congress of the International Federation of Automatic Control
  Barcelona, 21–26 July 2002 
EVOLUTIONARY GUIDANCE FOR MULTIPLE MISSILES
Evan J. Hughes*
* Department of Aerospace, Power and Sensors,
Cranfield University,
Royal Military College of Science,
Shrivenham, Swindon, UK, SN6 8LA.
ejhughes@iee.org

A new approach to guiding a swarm of missiles using an on-line evolutionary algorithm to optimise the flight paths is introduced. The missiles are flown via intermediate points which are adjusted dynamically by the evolutionary algorithm, then towards the point of impact with the target, thus altering the trajectory shape. Evolutionary algorithms are robust global optimisation techniques and are able to cope well with noise and uncertainty. This framework can be used to generate optimal flight profiles that satisfy multiple objectives and constraints. Example trajectories are presented for a multiple missile scenario using a highly non-linear model of a boost-glide missile.
Keywords: Data Fusion, Genetic Algorithms, Guidance Systems, Missiles, Multi-objective Optimisation
Session slot T-Tu-E06: Advances in Missile Guidance and Control/Area code 8a : Aerospace