Inverse Reinforcement Learning for Identifcation in Linear-Quadratic Dynamic Games

摘要

The theory of dynamic games has received considerable attention in a wide range of felds. While great efort has been made to develop new algorithms for fnding Nash equilibria in dynamic games, the identifcation of cost functions has received little attention. We present an identifcation algorithm for linear quadratic dynamic games, a framework which can be applied in the feld of shared control between a human and an automatic controller. In this application, the cost function describing human behavior is identifed, taking into account the infuence of the automation. Furthermore, we consider that human movement underlies certain variability by using a probabilistic Inverse Reinforcement Learning approach. As identifcation is performed in a single optimization step, the proposed method is suited for real-time applications. A simulation example shows that the algorithm successfully identifes the cost function of the frst player which—in combination with the second player—reproduces the observed system output.