面向吸气式高超声速飞行器弹性体的精确姿态控制研究

The safety of hypersonic flight is greatly affected by the strong coupling between elastic deformation and rigid modes, and the unstable zero dynamics caused by the non-minimum-phase system. Meanwhile, to ensure good scramjet inlet environment, the rigorous requirement is raised on precise attitude control that the dynamic error should be made sure to fall in ±5°. To deal with the above-mentioned problems, toward achieving fast adaptation and prescribed attitude performance, this project focused on the controller design for unstable flexible non-minimum-phase hypersonic flight vehicle. Firstly, the specific design for unstable zero dynamics caused by non-minimum-phase system will be presented to ensure the flight safety. Secondly, to fulfill the fast adaption of flying with hypersonic speed, the composite learning algorithm will be proposed by incorporating information of modeling error or disturbance estimation. Thirdly, the prescribed performance attitude control will be studied to guarantee the transient response to fulfill good scramjet inlet environment. The proposed adaption laws and controllers will be verified via simulation platform. Through this project, it is supposed to provide new progress for control of hypersonic flight vehicle with flexible effects and non-minimum phase to enhance flight safety and guarantee precise control performance.

弹性形变与刚体模态之间存在的动力学强耦合以及非最小相位特性带来的不稳定零动态使得高超声速飞行的安全性受到极大影响，同时动力装置超燃冲压发动机良好进气对精确控制提出高要求，需要姿态动态误差小于±5°。本项目针对不稳定高超声速飞行器研究非最小相位弹性体控制，实现智能快速自适应学习以及姿态预设性能跟踪。首先，考虑非最小相位特征实现不稳定零动态的有效控制；其次，利用建模误差以及扰动观测等实现不确定情形下的高超声速飞行快速自适应学习；再次，考虑超燃冲压发动机工作需求研究姿态预设性能控制，实现所需姿态动态响应；最终，通过仿真验证所给出的控制器设计方法。通过该项目研究，将在静不稳定弹性高超声速飞行器的安全飞行以及精确控制方面取得一定进展。

该项目面向高超声速飞行器动力特点、环境特点、飞行器外形特点，开展了高精度姿态控制、弹性体鲁棒控制以及非最小相位控制等一系列问题研究。.（1）针对弹性体存在的模态耦合开展了基于奇异摄动分解的动力学分解，具体分析系统时标特性进行系统分解获取快慢子系统；针对快系统设计滑模控制，针对慢系统设计滑模控制，最终进行控制合成实现弹性体的稳定控制。.（2）针对非最小相位特性开展了基于输出重定义的鲁棒切换控制，通过系统相对阶分析给出了重定义设计，得到输入输出子系统，并根据动力学特征开展了状态监测的鲁棒滑模智能控制设计保证了系统的快速收敛和精细调整。.（3）针对超燃冲压发动机对姿态控制的要求，开展了攻角约束的下的鲁棒设计，具体考虑攻角的约束特征通过虚拟控制量限制以及误差鲁棒控制保证攻角控制在预定范围内实现了高精度姿态控制，保证了系统任务的有效完成。.通过该项目研究发表9篇SCI论文，包括IEEE神经网络汇刊、IEEE SMC：Systems汇刊等；获得授权发明专利3项。