Combat survivability is an important objective in military air operations, which involves not being shot down by e.g. enemy aircraft. This involves analyzing data and information, detecting and estimating threats, and implementing actions to counteract threats. Beyond visual range missiles can today be fired from one hundred kilometers away. At such distances, missiles are difficult to detect and track. The use of techniques for recognizing hostile aircraft behaviors can possibly be used to infer the presence and for providing early warnings of such threats. In this paper we compare the use of dynamic Bayesian networks and fuzzy logic for detecting hostile aircraft behaviors.
This research has been supported by the Infofusion Re-search Program (University of Skövde, Sweden) in partnership with Saab AB and the Swedish Knowledge Foundation under grant 2010/0230 (UMIF). I would like to acknowledge Per-Johan Nordlund at Saab AB for providing simulated data, discussions and descriptions.