基于扰动观测器的指数时变滑模再入姿态控制

针对再入飞行器姿态控制问题，设计了一种全局鲁棒的基于扰动观测器的指数时变滑模姿态控制器，使得在存在匹配模型参数不确定性和外部扰动情况下仍能够实现精确的姿态控制。首先对飞行器模型进行反馈线性化解耦，将俯仰、滚转、偏航通道分解为3个独立的二阶系统。然后以俯仰通道为例，设计指数时变滑模控制器，使得系统具有全局鲁棒性。进一步，为了减小采用饱和函数消除滑模控制抖振而带来的系统跟踪误差，引入扰动观测器，设计了一种基于扰动观测器的指数时变滑模控制系统。最后，以某飞行器为例建立六自由度仿真模型，在考虑大气密度、转动惯量偏差以及外部高频扰动的情况下，验证了加入扰动观测器后对系统性能的提高。

Disturbance Observer Based Exponential Time-varying Sliding Mode for Re-entry Attitude Control

For re-entry vehicle attitude control problem, a disturbance observer-based exponential time-varying sliding mode attitude controller was designed to achieve precise attitude control when the matched parametric uncertainties and external disturbances occured. Firstly, feedback linearization method was applied to the nonlinear model of re-entry vehicle, which decomposed the pitch, roll and yaw channel decomposed into three independent second order system. Then, as an example, an exponential time-varying sliding mode controller (ETVSMC) was designed for the pitch channel. Furthermore, to reduce the static error caused by boundary layer solution which was used to soften the chattering, a disturbance observer based exponential time-varying sliding mode control (DOETVSMC) method was presented. Finally, a six degree-of-freedom simulation mode was established for a re-entry vehicle considering the atmospheric density, inertial uncertainties and external high-frequency perturbations, to demonstrate the improved performance of DOETVSMC with ETVSMC.