Abstract: In this article, a novel model-free dynamic inversion-based Q-learning (DIQL) algorithm is proposed to solve the optimal tracking control (OTC) problem of unknown nonlinear input-affine ...