Learning an Approximate Model Predictive Controller With Guarantees

By Michael Hertneck, Johannes Köhler, Sebastian Trimpe, Frank Allgöwer. IEEE Control Systems Letters, 2018

299 citations recorded by the source corpus.

A supervised learning framework is proposed to approximate a model predictive controller (MPC) with reduced computational complexity and guarantees on stability and constraint satisfaction. The framework can be used for a wide class of nonlinear systems. Any standard supervised learning technique (e.g., neural networks) can be employed to approximate the MPC from samples. In order to obtain closed-loop guarantees for the learned MPC, a robust MPC design is combined with statistical learning bounds. The MPC design ensures robustness to inaccurate inputs within given bounds, and Hoeffding’s Inequality is used to validate that the learned MPC satisfies these bounds with high confidence. The result is a closed-loop statistical guarantee on stability and constraint satisfaction for the learned MPC. The proposed learning-based MPC framework is illustrated on a nonlinear benchmark problem, for which we learn a neural network controller with guarantees.

Subfields: Control and Systems Engineering, Statistical and Nonlinear Physics

View Paper: https://www.semanticscholar.org/paper/63c2a93e17aea0808417526e704465d6994cb929

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Control and Systems Engineering

In Statistical and Nonlinear Physics

Citations in the corpus

Review on model predictive control: an engineering perspective (2021, background): “[45] took this thought focusing on accurate learning of the solution space by the neural network (NN).”
On the Sample Complexity of Stability Constrained Imitation Learning (2021, background): “, 2017), and can lead to policies that are more computationally efficient to evaluate online (Hertneck et al., 2018; Yin et al., 2020).”