摘要
针对近空间可变翼飞行器在小翼切换时存在参数不确定性的问题,本文设计了基于矩阵下三角对称(LDS)分解的小翼伸缩自适应切换控制器,以实现小翼伸缩的平滑切换。对近空间可变翼飞行器的非线性模型进行线性化处理,建立含有小翼伸缩变化的不确定线性模型;依据线性化的近空间可变翼飞行器模型,设计状态反馈控制器,并确定参考模型。然后设计自适应补偿控制器,消除参数不确定性的影响。将小翼切换自适应控制问题转换为参数不确定性问题,并且从理论上证明了小翼切换过程的稳定性。研究结果表明:提出的基于矩阵分解的自适应控制方法能够有效地解决小翼切换时飞行器参数不确定性问题,该方法有较强的鲁棒性,为飞行器参数不确定性问题提供了可靠地解决思路。
Considering the problem of parameter uncertainties in a near space morphing vehicle′s (NMV) winglet telescopic mode switching process, a winglet switching adaptive controller based on matrix lower triangular-diagonal symmetric(LDS) factorization is designed to realize the smooth switching between different modes. First, the NMV nonlinear model is linearized to establish an uncertain linear model with winglet switching. Based on the linearized model of the NMV, the state feedback controller is then designed and the reference model is determined. Next, the adaptive compensation controller is designed to eliminate the influence of parameter uncertainty. The problem of adaptive control for winglet switching is transformed into the problem of parameter uncertainties, and the stability of the winglet switching process is proven theoretically. Results show that the proposed adaptive control method based on matrix decomposition can effectively solve the parameters uncertainty of a NMV. This method has strong robustness and provides a reliable solution to the problem of parameter uncertainty of NVM.
作者
杨政
甄子洋
蒋烁莹
陶钢
YANG Zheng;ZHEN Ziyang;JIANG Shuoying;TAO Gang(College of Automation Engineering,Nanjing University of Aeronautics and Astronautics,Nanjing 210016,China;School of Engineering and Applied Science,University of Virginia,Charlottesville 22904-4743,USA)
出处
《哈尔滨工程大学学报》
EI
CAS
CSCD
北大核心
2019年第5期886-891,共6页
Journal of Harbin Engineering University
基金
国家自然科学基金项目(61304223
61673209
61741313)
航空科学基金(2016ZA52009)
一院高校联合创新基金(CALT201603)
南京航空航天大学研究生创新基地(实验室)开放基金项目(kfjj20170306)
关键词
近空间可变翼飞行器
小翼切换
状态反馈控制
自适应律
参考模型
参数不确定
矩阵分解
near space morphing vehicle (NMV)
winglet switch
state feedback control
adaptive update law
reference model
parameter uncertainties
matrix factorization