An Observer-Based Reinforcement Learning Solution for Model-Following Problems
In this paper, a multi-objective model-following control problem is solved using an observer-based adaptive learning scheme. The overall goal is to regulate the model-following error dynamics along with optimizing the dynamic variables of a process in a model-free fashion. This solution employs an integral reinforcement learning approach to adapt three strategies. The first strategy observes the states of desired process dynamics, while the second one stabilizes and optimizes the closed-loop system. The third strategy allows the process to follow a desired reference-trajectory. The adaptive learning scheme is implemented using an approximate projection estimation approach under mild conditions about the learning parameters.
PDF Abstract
