Unmanned aerial vehicle air combat maneuver decision-making method based on distance-first experience playback

Abstract

The invention discloses an unmanned aerial vehicle air combat maneuver decision-making method based on a distance priority constraint experience playback method. The method comprises an unmanned aerial vehicle air combat maneuver simulation preparation information setting step, an unmanned aerial vehicle air combat maneuver decision-making agent model building step, an unmanned aerial vehicle air combat maneuver decision-making agent model training step and a multi-time training and ending step. Compared with an existing priority experience playback method, the method has the advantages that the calculation logic for calculating the sample priority is improved, and the distance priority is introduced to relieve the inaccuracy of TD-error in the initial training stage, so that a sample close to a termination state is preferentially selected during intelligent agent learning in the initial training stage; therefore, meaningless learning of the intelligent agent at the initial stage of training is avoided, the training efficiency and the sample utilization rate are greatly improved, and the time cost of training is reduced.

Description

technical field [0001] The present invention relates to the field of virtual simulation of UAV air combat, in particular, it relates to a UAV air combat maneuver decision-making method based on distance-priority experience playback, which can use deep reinforcement learning methods to speed up the development of UAV air combat maneuver decision-making intelligent body models. The training in the air combat simulation, on the basis of the traditional priority experience playback method, further improves the sample utilization efficiency, avoids the meaningless learning of the UAV air combat maneuver decision-making agent model in the early stage of training, and improves the UAV air combat maneuver decision-making. The speed at which the agent model completes air combat maneuver training. Background technique [0002] With the development of unmanned and intelligent technology, the use of drones has become an important topic in the fields of civil and military science. The in...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to propose a UAV air combat maneuver decision-making method based on distance priority experience playback. Compared with the existing priority experience playback method, the key improvement is to use distance priority to solve the priority based on TD-error. There is a large deviation in the early stage of training, and UAV air combat maneuver samples are selected for training times to attenuate the influence of distance priority on the total priority in the middle and late training, so as to avoid UAV air combat maneuver decision-making agents The model does a lot of meaningless learning in the early stage of training, which improves training efficiency and sample utilization

Images