控制理论(社会学)
非完整系统
参数统计
格拉米安矩阵
数学
李雅普诺夫函数
计算机科学
跟踪(教育)
李雅普诺夫方程
控制(管理)
线性系统
控制系统
可控性
线性控制系统
理论(学习稳定性)
应用数学
可控性
李雅普诺夫指数
控制工程
标识
DOI:10.1080/00207721.2026.2711939
摘要
This paper investigates the control and observation problems for discrete-time linear time-varying systems. By combining finite-horizon controllability/reconstructibility Gramians with difference parametric Lyapunov equations, new state-feedback and observer design methods are developed. Under suitable finite-horizon Gramian conditions, explicit feedback and observer gains are constructed such that the resulting closed-loop systems are asymptotically stable or uniformly exponentially stable. In addition, boundedness properties of the associated matrix sequences and gains are established. To further reveal the structure of the proposed designs, two classical alternative gain formulas are revisited and further characterised, and precise relationships between the original gains and the alternative ones are derived. These results provide additional insight into the role of the finite-horizon parameters in controller and observer synthesis. The state-feedback and observer designs are then combined to address observer-based output-feedback stabilisation. Sufficient conditions are derived to guarantee the uniform exponential stability of the resulting closed-loop system. Finally, the proposed theory is applied to the trajectory tracking problem of a class of nonholonomic systems.
科研通智能强力驱动
Strongly Powered by AbleSci AI