A method and device for missile guidance based on reinforcement learning

Abstract

The present invention provides a missile guidance method and device based on reinforcement learning, wherein the method includes: based on radar signals, acquiring the position and attitude information of the current missile and the target, and the current remaining energy information of the missile; based on the position and attitude of the current missile and the target Information, as well as the current remaining energy information of the missile, determine the observation vector at the current moment, and determine the current state based on the observation vector at the current moment and the previous moment; input the current state to the action network, and obtain the action output by the action network , and based on the action, determine the control amount of the missile; wherein, the action network is obtained by reinforcement learning based on the sample observation vector, the sample action taken by the missile at each moment and its corresponding reward. The invention improves the accuracy of rounding up missiles with angle constraints.

Description

technical field [0001] The invention relates to the technical field of missile guidance, in particular to a missile guidance method and device based on reinforcement learning. Background technique [0002] In recent years, the issue of missile guidance has received extensive attention. Due to the improvement of the capabilities of modern aircraft, the flight trajectory of the aircraft is changeable, and the breakthrough ability is enhanced, which puts forward higher requirements for the missile guidance control method. In addition, because the missile has complex dynamic characteristics, unstable flight environment, and high control precision, the control system needs to be robust and adapt to changing targets and environments. Traditional control methods can only be adapted to certain types of tasks. However, in the process of modern aircraft penetration, the relative relationship between missiles and target aircraft may change rapidly, making traditional control methods d...

AI Technical Summary

Problems solved by technology

[0004] The present invention provides a missile guidance method and device based on reinforcement learning, which is used to solve the defect that it is difficult to realize the round-up of missiles with angle constraints in the prior art

Images