BTT导弹的神经网络自适应反馈线性化控制

 提出基于反馈线性化理论的BTT导弹自动驾驶仪设计方法。并提出由俯仰过载命令产生攻角命令的方法。应用CMAC神经网络构成一种自适应反馈线性化方案,使具有不确定性的实际系统获得要求的跟踪性能。闭环系统的稳定性由李雅普诺夫稳定性理论得到保证。该方法适合于控制导弹作全空域飞行。

NEURAL NETWORKS BASED ADAPTIVE FEEDBACK LINEARIZATION FOR BANK-TO-TURN MISSILE CONTROL

Based on the theory of feedback linearization and the simplified bank to turn (BTT) control design model, a BTT missile autopilot design method is presented. No zero dynamics exists in the proposed scheme. A method for generating an attack command from the normal over load command is presented. Based on this, the CMAC neural networks are introduced for adaptive feedback linearization of bank to turn missiles, and hence the desired tracking performance of the practical missile system with uncertainty is achieved. A stability proof is given strictly in the sense of Lyapunov. The proposed scheme is fit for the entire flight envelope of the BTT missile. Simulation results have shown the rightness and effectiveness of the proposed scheme.