Modeling of Air-to-air Missile Dynamic Attack Zone Based on Bayesian Networks

In the beyond visual range (BVR) air combat, the low hit rate of air-to-air missile (AAM) caused by evasive maneuvers of targets has long been a question of great interest. This paper attempts to address this problem by sending a corrected relay guidance command based on the missile dynamic attack zone (DAZ) to adjust the trajectory of the missile in real time. First, the physical model of the attack zone (AZ) is built and analyzed with the conditions of maneuvering target. Second, considering the increasing complexity of the environment, large battlefield range and the high mobility of targets, the traditional AZ can not meet the needs of current BVR air combat very well. Thus, to adapt the changes of air combat better, the probabilistic dependencies between multiple kinematic parameters and DAZ is modelled by Bayesian Networks (BN). The conditional probability tables (CPT) of BN is learnt from datasets derived from the physical model. Finally, through the inference and analysis of BN, we get an additional relay guidance command to adjust the trajectory of the missile in real time. The proposed method is exemplified by simulation, and results show that this method is effective and has a higher hitting accuracy.