基于观测器的不匹配未知输入系统的滑模控制

对具有不匹配时变未知输入的线性系统，研究了基于观测器的滑模控制器设计方法.首先，设计了一种与控制输入无关、能直接抵消未知输入影响的降维观测器.通过该观测器估计系统状态，并提出了一种未知输入代数重构方法，进而利用高增益滑模观测器给出了未知输入一阶微分的渐近估计方法.最后，提出了一种基于系统状态估计、未知输入重构及未知输入一阶微分估计滑模控制器新的设计方法，并通过仿真分析，论证了该方法的可行性和正确性.
An investigation was made into the sliding mode control design method for a class of linear system with unmatched time varying unknown inputs. A reduced order observer was developed to be independent of the control inputs and able to overcome the influence of the unknown inputs. Moreover, an algebraic unknown input reconstruction method based on state estimation was developed. Besides, an estimation method of the first order derivatives of the unknown inputs was proposed by means of a high gain sliding mode observer. Finally, based on the estimation of the system states, the reconstruction of the unknown input and the first derivative estimation, a novel sliding mode controller was constructed，and a simulation analysis validated the effectiveness of the proposed method