Direct Learning for Parameter-Varying Feedforward Control: A Neural-Network Approach

TL;DR

This paper introduces a neural network-based method for learning parameter-varying feedforward controllers that adapt to system dynamics, enabling effective control of LPV systems through data-driven optimization.

Contribution

It develops an input-output LPV feedforward parameterization with neural networks and a Levenberg-Marquardt optimization approach for efficient data-driven controller learning.

Findings

Demonstrates excellent performance in simulation of LPV systems.

Enables compensation of a wide class of LPV systems.

Provides an analytic gradient-based optimization method.

Abstract

The performance of a feedforward controller is primarily determined by the extent to which it can capture the relevant dynamics of a system. The aim of this paper is to develop an input-output linear parameter-varying (LPV) feedforward parameterization and a corresponding data-driven estimation method in which the dependency of the coefficients on the scheduling signal are learned by a neural network. The use of a neural network enables the parameterization to compensate a wide class of constant relative degree LPV systems. Efficient optimization of the neural-network-based controller is achieved through a Levenberg-Marquardt approach with analytic gradients and a pseudolinear approach generalizing Sanathanan-Koerner to the LPV case. The performance of the developed feedforward learning method is validated in a simulation study of an LPV system showing excellent performance.