A Robust LQTI-DOBC Framework for Disturbance Rejection and Command Tracking Using Integral Reinforcement Learning

This paper proposes a robust tracking control framework that integrates Disturbance Observer-Based Control (DOBC) with Linear Quadratic Tracker with Integrator (LQTI) using Integral Reinforcement Learning (IRL). Traditional control methods often struggle to ensure robust performance when external disturbances change rapidly or model uncertainties are significant, potentially leading to degraded command tracking performance. To address this challenge, the proposed framework combines the LQTI structure to enhance command tracking performance with IRL for real-time gain adaptation and DOBC for effective disturbance estimation, thereby simultaneously improving disturbance rejection and command tracking capabilities. The integrated control system is designed and validated through numerical simulations in a linear time-invariant system under dynamic disturbance conditions.