Adaptive Augmentation of a New Baseline Control Architecture for Tail-Controlled Missiles Using a N onlinear Reference Model *
Adaptive Augmentation of a New Baseline Control Architecture for Tail-Controlled Missiles Using a N onlinear Reference Model *
复制标题
使用非线性参考模型自适应增强尾控导弹的新基线控制架构 *
DOI：
--
复制
发表时间：
2012
期刊：
影响因子：
--
通讯作者：
F. Holzapfel
中科院分区：
文献类型：
--
作者：
F. Peter;M. Leitão;F. Holzapfel
一种基于非线性参考模型的自适应增强非线性动态反转（NDI）控制结构是为高度敏捷、尾部控制的地对空导弹开发的。由于采用模块化控制设计，导弹能够执行滑行转弯（STT）和转弯时倾斜（BWT）机动。本文的重点在于非线性参考模型以及NDI误差反馈架构的相应变化。为了克服尾控导弹的非最小相位特性，对俯仰加速度和偏航加速度进行转换，使系统具有最小相位。开发了一个具有导弹动力学所有主要非线性效应的非线性参考模型。通过这种非线性命令滤波器，闭环系统可以充分利用工厂的物理能力。基于经典的二环反演策略，NDI误差反馈基线控制器的设计根据参考模型的修改进行调整。通过利用 NDI，跟踪问题转化为具有几乎线性误差动态的稳定问题。这使得系统非常适合使用模型参考自适应控制（MRAC）方法作为自适应层，能够应对建模误差和传感器偏差。自动驾驶仪设计使用蒙特卡罗模拟证明了其在大量不确定性组合下的鲁棒性和性能。
An adaptive augmented Nonlinear Dynamic Inversion (NDI) control structure based on a nonlinear reference model is developed for a highly agile, tail-controlled surface-to-air missile. Due to a modular control design, the missi le is able to perform skid-to-turn (STT) and bank-while-turn (BWT) maneuvers. The focus of this paper lies on the nonlinear reference model and the respective change in the NDI error feedback architecture. In order to overcome the non-minimum phase property of tail-controlled missiles, a conversion of the pitch- and yaw-acceleration, which renders the syst em minimum phase is conducted. A nonlinear reference model featuring all the main nonlinear effects of the missile dynamics is developed. With such a nonlinear command filter, the closed-loop system fully exploits the physical capabilities of the plant. Based on a clas sical two-loop inversion strategy, the design of the NDI error feedback baseline controller is ad justed in accordance to the modification of the reference model. By making use of NDI, the t racking problem is transformed into a stabilizing problem with an almost linear error dyn amics. This makes the system perfectly suitable for using a Model Reference Adaptive Control (MRAC) approach as an adaptive layer, which is able to cope with modeling errors a nd sensor biases. The autopilot design proved its robustness and performance under a large spectrum of uncertainty combinations using Monte Carlo simulations.