Neural Data-Enabled Predictive Control

TL;DR

This paper introduces neural network-based data-enabled predictive control (DeePC) for nonlinear systems, leveraging neural basis functions to improve trajectory prediction and control, with methods for consistency and computational efficiency.

Contribution

It develops neural DeePC algorithms that use deep neural networks to learn neural bases for nonlinear systems, enabling online affine interpolation for control.

Findings

Neural DeePC effectively controls a nonlinear pendulum.

The approach improves prediction accuracy over traditional methods.

Methods for ensuring consistency and reducing complexity are proposed.

Abstract

Data-enabled predictive control (DeePC) for linear systems utilizes data matrices of recorded trajectories to directly predict new system trajectories, which is very appealing for real-life applications. In this paper we leverage the universal approximation properties of neural networks (NNs) to develop neural DeePC algorithms for nonlinear systems. Firstly, we point out that the outputs of the last hidden layer of a deep NN implicitly construct a basis in a so-called neural (feature) space, while the output linear layer performs affine interpolation in the neural space. As such, we can train off-line a deep NN using large data sets of trajectories to learn the neural basis and compute on-line a suitable affine interpolation using DeePC. Secondly, methods for guaranteeing consistency of neural DeePC and for reducing computational complexity are developed. Several neural DeePC formulations are illustrated on a nonlinear pendulum example.