Effectiveness of Autonomous Decision Making for Unmanned Combat Aerial Vehicles in Dogfight Engagements

A study is presented which shows that unmanned combat aerial vehicles (UCAVs) are suitable for dogfighting (DF) engagements, and they can autonomously perform aggressive DF maneuvers. In the context of autonomous decision making by the UCAV, the max–min search algorithm is a practical and effective method for solving DF games especially when the opponent's level of intelligence (LOI) is set to high or medium. When the opponent's LOI is set to low, less-conservative strategies could be more suitable to make the most of blue's advantage. Also, instead of relying on stereotyped (prestored) maneuvers, the use of a continuous guidance law based on lead/pure/lag pursuit and climb/dive maneuvers is a natural and effective solution. Such a continuous guidance law is optimized and must be discretized to apply the game-theoretic max–min search. The higher the resolution of that discretization is used, the better approximation is obtained, but the computational load is proportional to the square of the number of samples, and so it necessary to make a design tradeoff between resolution and computational load.