Missile longitudinal attitude control algorithm based on reinforcement learning

Abstract

The invention provides a missile longitudinal attitude control algorithm based on reinforcement learning, belongs to the field of missile attitude control research, and relates to an Actor-Critic (AC)structure based on reinforcement learning, which is composed of an action network and an evaluation network, wherein the evaluation network is used for outputting an evaluation value for the state ofthe missile according to the state of the missile, and the action network is used for generating a corresponding elevator deflection angle according to the evaluation value output by the evaluation network, so that the longitudinal attitude of the missile is stably controlled under the condition of not depending on an internal model of the missile. The method comprises the following steps: 1) establishing and determining a missile longitudinal attitude kinetic model; (2) defining a tracking error of the missile attack angle and establishing a performance index related to the error; (3) designing an evaluation network; (4) designing an action network; (5) designing an evaluation network weight updating law; (6) designing an action network weight updating law. The algorithm is mainly applied to the missile longitudinal attitude control.

Description

technical field [0001] The invention relates to a missile longitudinal attitude control algorithm based on reinforcement learning, which belongs to the research field of missile attitude control. Background technique [0002] Aerospace vehicles have received more and more attention in recent years. Due to the expansion of the flight envelope of the aircraft, more and more tasks are required for the aircraft, which means that the flight control of the aircraft is becoming more and more complicated. Due to the complex dynamic characteristics of the missile, the uncertain flight environment and high control precision are required. The control system needs to be robust and adaptive to modeling uncertainties. Based on the above requirements, traditional control methods have been difficult to handle many complex control tasks. [0003] The most widely used in the field of missile attitude control is the PID controller. The PID controller has the characteristics of simple struct...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to provide a method for controlling the longitudinal attitude of missiles with modeling uncertainty and model parameters that cannot be accurately obtained while ensuring a simple structure of the control law. Missile longitudinal attitude control algorithm based on reinforcement learning to stabilize the missile's longitudinal attitude angle

Images