Title :
Engagement maneuvering strategy of air combat based on fuzzy markov game theory
Author :
An, Xu ; Yingxin, Kou ; Lei, Yu ; Baowei, Xu ; Yue, Lv
Author_Institution :
Eng. Coll., Air Force Eng. Univ., Xi´´an, China
Abstract :
Based on the framework of markov game theory, the attack engagement maneuvering strategy of 1 vs. 1 air combat is studied. First, the markov game model of air combat of two fighters is established with an introduction of its state space description and action set, and the maneuvering target set of both rivals is established. Second, the fuzzy theory is utilized and a fuzzy Markov model framework is introduced with its fuzzy state space and fuzzy reward function. Then, with the fuzzy markov game, the Nash-Q learning method is adopted to obtain the optimized strategy of the fighters. At last, a simulation shows that with the method in this paper, the 1 vs. 1 air combat decision making can be solved and the method is effective.
Keywords :
Markov processes; fuzzy set theory; game theory; learning (artificial intelligence); military systems; multi-agent systems; Nash-Q learning method; air combat; attack engagement maneuvering strategy; engagement maneuvering strategy; fuzzy Markov game theory; fuzzy reward function; fuzzy state space; fuzzy theory; state space description; Atmospheric modeling; Game theory; Games; Joints; Learning; Markov processes; Air Combat Decision Making; Fuzzy State Space; Markov Game; Multi-agent System; Reinforcement Learning;
Conference_Titel :
Computing, Control and Industrial Engineering (CCIE), 2011 IEEE 2nd International Conference on
Conference_Location :
Wuhan
Print_ISBN :
978-1-4244-9599-3
DOI :
10.1109/CCIENG.2011.6008083