基于神经网络和扰动观测的综合火力/飞行智能控制方法OA

Considering the control instability caused by highly dynamic trajectory tracking and attack window compression,this paper proposes an intelligent learning-based integrated fire/flight control(IFFC)method.The fire tracking controller is constructed based on the relative motion relationship between the aircraft and the target.The adaptive fuzzy control strategy is employed to resolve guidance commands online,achieving asymptotic convergence of aiming errors through dynamic adjustment of membership functions.The six-degree-of-freedom dynamic model is decoupled into the angle-of-attack,sideslip angle,and bank angle subsystems.For these subsystems,neural network composite learning laws and disturbance observers are developed,while adaptive control laws are designed to obtain desired control moments.The optimization allocation model for aerodynamic control surfaces and thrust vectoring is established,with the sequential quadratic programming(SQP)algorithm applied to solve optimal control allocation online.Simulation results demonstrate that the proposed method achieves rapid convergence of aiming errors while exhibiting enhanced learning performance and high-precision command tracking.