Imitation Learning from Nonlinear MPC via the Exact Q-Loss and its Gauss-Newton Approximation

Open in new window