滚仰式半捷联稳定平台的神经网络自适应控制算法

A Neural Network Adaptive Control Algorithm for Roll-Pitch Semi-Strapdown Stabilized Platform

基于滚仰式半捷联稳定平台的机械结构特性,建立了稳定平台的运动学及动力学模型。为了解决动力学建模不确定性以及摩擦力等外界干扰对稳定平台工作性能的影响,实现滚仰式导引头的捷联稳定控制,设计了一种神经网络自适应控制算法,并将其应用于稳定平台的控制之中。仿真结果表明,相比传统方法,提出的算法可以有效提高滚仰式稳定平台抵抗外界干扰的能力,在动力学建模误差、摩擦等扰动很大的情况下仍可以保证滚仰式稳定平台的控制精度,其结果可以为滚仰式导引头的稳定与跟踪控制提供研究基础。

Based on the mechanical structure of the roll-pitch platform, the kinematics model and the dynamic models were established. In order to eliminate the effects of the external disturbances such as dynamic modeling uncertainty and friction on the stabilized platform performance, and achieve the strapdown stabilization control of the roll-pitch seeker, a neural network adaptive control algorithm was designed and applied to the control of a stabilized platform. The simulation results show that compared with the conventional control methods, the proposed control algorithm can effectively improve the control performance of the roll-pitch stabilized platform, the tracking ability can be ensured under the external disturbances such as modeling error and friction. These results can serve as a basis for the control of roll-pitch stabilized platform.