An application of machine learning techniques to the automatic acquisition from experience of tactical expertise in multiaircraft combat

The authors describe an application of machine learning techniques to the acquisition of opponent allocation rules in multiaircraft air combat. They outline the tools used: the multiaircraft combat simulator EMIL and the learning system MACHIN. Then they explain how they were integrated in EMILIA and discuss some results of the first test and validation campaigns. The problems concerning the inclusion of learning capabilities in the air combat simulation environment and the solutions implemented are presented. It was found that the learning techniques implemented already allow operationally valuable rule bases to be created and included in combat situations. They allow a refinement of the expertise in the area of two-to-two multiaircraft combat as well as one-to-two asymmetrical combat