Strategy Synthesis in Aerial Dogfight Game Model.
Abstract
The main problem of interest in the report is the role-definition problem arising in one-on-one dogfight game models. The computational approach is aimed at providing a decomposition of the space of game initial conditions into sets of unilateral capture capability for each of the players, and at outlining the draw and sacrifice sets in accordance with the players' individual preferences for game outcomes. The procedure develops the feedback policy (in terms of the observable data) that attains the above decomposition. Two highly simplified one-on-one games are considered. The first game model is a discrete time-state alternating move game (perfect information) on a horizontal grid reminiscent of the Isaacs examples. The second model is a continuous time-regional feedback game (imperfect information) in the horizontal plane. (Modified author abstract)
Document Details
- Document Type
- Technical Report
- Publication Date
- Mar 01, 1973
- Accession Number
- AD0764352
Entities
People
- Arthur Kaercher
- Gilbert Carpenter
- Michael Falco
- Victor Cohen
Organizations
- Grumman