不变结构半弹道式再入飞行器的建模与多模型方法估计

摘要

以配平攻角状态再入的不变结构半弹道式再入飞行器(Semi-ballistic reentry vehicle,SBRV)不同于传统的弹道式再入飞行器(Ballistic reentry vehicle,BRV)和机动再入飞行器(Maneuvering reentry vehicle,MaRV),本文分析了该飞行器再入特征,提出了新的模型并分析了该模型与传统再入模型间的关系.对该再入问题的多模型混合状态估计器引入了 F-均匀模型集与期望模式补偿(Expected-mode augmentation,EMA)集.根据SBRV的圆柱体状模式空间的需求,文中扩展了现有的方法以设计F-均匀模型集,进而提出一种EMA集的实现形式.前者在分布最小失配意义下使估计器最优;后者相比于前者具有更高的估计精度.仿真结果表明,相比于传统Monte-Carlo法生成的模型集,在模型集势相当的情况下这两种模型集对不变结构SBRV再入的初始阶段有更高的模式估计的精度,在该飞行器状态变化剧烈时有更高的混合状态估计精度.%A trim flying semi-ballistic reentry vehicle (SBRV), with invariable structure during reentry process, is different from the conventional ballistic reentry vehicle (BRV) or the maneuvering reentry vehicle (MaRV). Characteristics of the vehicle are analyzed in this paper, then the SBRV model is proposed and compared to the conventional models used for reentry vehicles. For the multiple-model hybrid state estimator of the reentry problem, the F-uniform model set and the expected-mode augmentation (EMA) set are applied. According to SBRV′s cylinder mode space, the present design method is expanded for the F-uniform model set, then an EMA approach is proposed. The former optimizes the estimator in the sense of minimum distribution mismatch and the latter has higher accuracy than the former. Simulation results show that the two model sets have higher accuracy in mode estimation for the initial part of the invariable-structure SBRV's reentry process and have higher hybrid state estimation accuracy when the vehicle's state changes greatly, as compared with the model set generated by the conventional Monte-Carlo method.