Study on Threat Assessment Based on Rough Ret and Support Vector Machines for Regression in BVR Air Combat

Abstract

Aiming at the shortage of existing TA methods in BVR air combat, a new TA method based on RS and SVM for regression is advanced, and advantage functions models are structured. Firstly, Missile scope is introduced as a factor of distance, so advantage function of distance is improved. Moreover, advantage functions such as azimuth, enter angle, energy, efficiency are optimized, and goal campaign value is considered as a factor of TA. Secondly, redundancy information of each advantage function is eliminated by reduction of knowledge based on RS, and the final taxis of TA are realized by SVM for regression. Finally, simulation results show the method is rational and effective.