Model Reference Adaptive Control with Input and Output Constraints

本文的研究目标是设计、分析并实现一种具有输入输出信号幅度和变化率约束的模型参考自适应控制算法。本文将考虑该算法对有界扰动和时变参数的鲁棒性。在开发高度复杂的自适应算法方面已经做了很多工作。然而，现有的结果在很大程度上忽略了现实投入和产出约束的存在。到目前为止，已经建立了具有饱和输入和开环稳定对象的自适应控制算法的稳定性结果。本文提出了一种基于时变参考模型、时变调节动力学和远期预测原理的模型参考自适应控制算法。使用这种方法，可以在很大程度上避免输入输出信号饱和导致的不稳定。其目的是获得良好的跟踪和调节性能，同时满足预定的输入和输出约束。所开发的算法将通过稳定和不稳定开环装置的仿真进行分析和广泛的验证。它将最终应用于人体血压的控制和机器人操纵器的控制。

The goal of the proposed research is to design, analyze and implement a model reference adaptive control algorithm with constraints on the amplitude and the rate of change of input and output signals. The robustness of the proposed algorithm with respect to bounded disturbance and time-varying parameters will be considered. Much work has been done in the development of highly sophisticated adaptive algorithms. The existing results, however, largely ignore the presence of realistic input and output constraints. So far the stability results have been established for adaptive control algorithms with saturating inputs and open-loop stable plants. In this work a model reference adaptive control algorithm is proposed with time-varying reference model, time-varying regulation dynamics and long-range prediction principle. Using this method the saturation of the input and output signals, which can lead to instability, is to a large extent avoided. The aim is to obtain good tracking and regulation properties satisfying at the same time prespecified input and output constraints. The developed algorithm will be analyzed and extensively checked by simulations for both stable and unstable open-loop plants. It will be finally applied for control of blood pressure in human body and for control of robotic manipulators.