引用本文:唐功友, 高德欣, 张宝琳.基于观测器的受扰非线性系统近似最优跟踪控制[J].控制理论与应用,2007,24(5):725~731.[点击复制]
TANG Gong-you, GAO De-xin, ZHANG Bao-lin.An observer-based approximate optimal tracking control for nonlinear systems with persistent disturbances[J].Control Theory and Technology,2007,24(5):725~731.[点击复制]
基于观测器的受扰非线性系统近似最优跟踪控制
An observer-based approximate optimal tracking control for nonlinear systems with persistent disturbances
摘要点击 1599  全文点击 1314  投稿时间：2005-05-30  修订日期：2006-06-13
查看全文  查看/发表评论  下载PDF阅读器
DOI编号  
  2007,24(5):725-731
中文关键词  非线性系统  扰动  最优控制  输出跟踪控制  降维观测器
英文关键词  nonlinear systems  disturbances  optimal control  output tracking control  reduced-order observer
基金项目  国家自然科学基金资助项目(60574023); 山东省自然科学重点基金资助项目(Z2005G01); 青岛市自然科学基金项目(05-1-JC-94).
作者单位
唐功友, 高德欣, 张宝琳 中国海洋大学信息科学与工程学院, 山东青岛266100
青岛科技大学自动化与电子工程学院, 山东青岛266042
中国计量学院数学与信息科学系, 浙江杭州310018 
中文摘要
      研究一类受扰非线性系统的最优输出跟踪控制问题. 给出了有限时域最优输出跟踪控制律的近似设计算法. 首先将求解受扰非线性系统最优跟踪控制问题转换为求解状态向量与伴随向量耦合的非线性两点边值问题,利用逐次逼近方法构造序列将其转化为求解两个解耦的线性微分方程序列问题. 通过迭代求解伴随向量的序列, 可得到由解析的线性前馈-反馈控制部分和伴随向量的极限形式的非线性补偿部分组成的最优输出跟踪控制律. 利用参考输入降维观测器和扰动降维观测器, 解决了前馈控制的物理可实现问题. 最后仿真结果表明了该方法的有效性.
英文摘要
      The optimal output tracking control (OOTC) problem for a class of nonlinear systems with persistent disturbances is considered. An approximate design algorithm of the OOTC law is presented in the finite time domain. Firstly, the two-point boundary value (TPBV) problem, which is derived from the original OOTC theory, is transformed to a coupled nonlinear TPBV problem in state vectors and adjoint vectors. The coupled TPBV problem is further transformed to two decoupled linear differential sequences via a recently developed successive approximation approach (SAA). By iteratively solving the adjoint vector sequence, the OOTC law can be obtained, which consists of analytic linear feedforward and feedback terms, as well as a nonlinear compensation term determined by the limit of the adjoint vector sequence. Furthermore, a reduced-order reference input observer and a reduced-order disturbance observer are constructed in order to solve the physically realizable problem of feedforward control. Finally, simulation examples show the effectiveness of the presented approach.