BP-MPC: Optimizing the Closed-Loop Performance of MPC using BackPropagation

TL;DR

This paper introduces BP-MPC, a backpropagation-based method to optimize MPC parameters for improved closed-loop performance, providing convergence guarantees and handling feasibility issues.

Contribution

It presents a novel backpropagation scheme for tuning MPC that accounts for nonlinear dynamics and guarantees convergence, addressing a key open problem.

Findings

Enables optimal tuning of MPC using backpropagation.

Handles nonlinear system dynamics and feasibility losses.

Provides convergence guarantees for the optimization process.

Abstract

Model predictive control (MPC) is pervasive in research and industry. However, designing the cost function and the constraints of the MPC to maximize closed-loop performance remains an open problem. To achieve optimal tuning, we propose a backpropagation scheme that solves a policy optimization problem with nonlinear system dynamics and MPC policies. We enforce the system dynamics using linearization and allow the MPC problem to contain elements that depend on the current system state and on past MPC solutions. Moreover, we propose a simple extension that can deal with losses of feasibility. Our approach, unlike other methods in the literature, enjoys convergence guarantees.