Minimax Q-learning control for linear systems using the Wasserstein metric

Stochastic optimal control usually requires an explicit dynamical model with probability distributions, which are difficult to obtain in practice. In this work, we consider the linear quadratic regulator (LQR) problem of unknown systems and adopt a Wasserstein penalty address distribution uncertainty additive stochastic disturbances. By constructing equivalent deterministic game penalized LQR problem, propose Q-learning method convergence guarantees learn minimax controller.