Research Article
Improving Maneuver Strategy in Air Combat by Alternate Freeze Games with a Deep Reinforcement Learning Algorithm
Algorithm 2
Alternate freeze game DQN for maneuver guidance agent training in air combats.
(1) | Set parameters of both aircrafts | (2) | Set simulation parameters | (3) | Set the number of training periods and the condition for ending each training period | (4) | Set DRL parameters | (5) | Set the opponent initialization policy | (6) | for period = 1, do | (7) | for aircraft = [red, blue] do | (8) | if aircraft = red then | (9) | Set the opponent policy | (10) | Initialize neural networks of red agent | (11) | while Winning rate< do | (12) | Train agent using Algorithm 1 | (13) | end while | (14) | Save the well-trained agent, whose maneuver guidance policy is | (15) | else | (16) | if period = then | (17) | break | (18) | else | (19) | Set the opponent policy | (20) | Initialize neural networks of blue agent | (21) | while Winning rate< do | (22) | Train agent using Algorithm 1 | (23) | end while | (24) | Save the well-trained agent, whose maneuver guidance policy is | (25) | end if | (26) | end if | (27) | end for | (28) | end for |
|