Learning Controllers for Unstable Linear Quadratic Regulators from a Single Trajectory
share
We present the first approach for learning – from a single trajectory – a linear quadratic regulator (LQR), even for unstable systems, without knowledge of the system dynamics and without requiring an initial stabilizing controller. Our central contribution is an efficient algorithm – eXploration – that quickly identifies a stabilizing controller. Our approach utilizes robust System Level Synthesis (SLS), and we prove that it succeeds in a constant number of iterations. Our approach can be used to initialize existing algorithms that require a stabilizing controller as input. When used in this way, it yields a method for learning LQRs from a single trajectory and even for unstable systems, while suffering at most 𝒪(√(T)) regret.
READ FULL TEXT
