SITUATION ASSESSMENT IN AIR-COMBAT: A FUZZY-BAYESIAN HYBRID APPROACH

In modern air combat operations, the mental workload for fighter pilots is extremely high. The pilot has to make fast dynamic decisions under high uncertainty and high time pressure. This is hard to perform in Within Visual Range (WVR) combat operations, but becomes even harder in Beyond Visual Range (BVR) combat operations where the on-board sensors of aircraft become the pilot's eyes and ears. Typically, the data received from multiple on-board and off-board sensors and sources is fused mentally by operators to produce a coherent air surveillance picture portraying tracks of airborne targets and their classification. Then the air surveillance picture is analyzed mentally to determine the behavior of each target with respect to the own ship and other targets in the region and assess the intent or threat that they pose or the impact they may have on the mission (situation assessment). As the number of targets grows or the situation escalates, the volume of available data from these sensors and sources may overload the operators. To assist them in such situations, it is desirable to automate some of the situation and threat assessment process. In this paper, a Fuzzy logic and Bayesian Network (BN) based hybrid technique is used to investigate the possibilities of design and implementation of an expert system named Intelligent System for Situation Assessment in Air-Combat (ISSAAC) as an aid to pilots engaged in air-combat. ISSAAC is a pilot-in-loop (PIL) simulator consisting of integration of platform models, sensor models, pilot mental models and data processing algorithms. The capability of ISSAAC is demonstrated by simulating an air-to-air combat scenario consisting of six targets