A physics-guided data-driven feedforward tracking controller for systems with unmodeled dynamics -- applied to 3D printing

TL;DR

This paper introduces a hybrid physics-guided data-driven feedforward controller that improves tracking accuracy and stability in systems with unmodeled dynamics, demonstrated on 3D printing applications.

Contribution

It proposes a novel hybrid FBF controller combining physics-based and data-driven models for enhanced tracking and stability in systems with unmodeled dynamics.

Findings

Significantly improved tracking accuracy in high-speed 3D printing.

Effective detection of potential instability during operation.

Robust handling of delays in data acquisition.

Abstract

A hybrid (i.e., physics-guided data-driven) feedforward tracking controller is proposed for systems with unmodeled linear or nonlinear dynamics. The controller is based on the filtered basis function (FBF) approach, hence it is called a hybrid FBF controller. It formulates the feedforward control input to a system as a linear combination of a set of basis functions whose coefficients are selected to minimize tracking errors. The basis functions are filtered using a combination of two linear models to predict and minimize the tracking errors. The first model is physics-based and remains unaltered during the execution of the controller, while the second is data-driven and is continuously updated during the execution of the controller. To ensure its practicality and safe learning, the proposed hybrid FBF controller is equipped with the ability to handle delays in data acquisition and to detect impending instability due to its inherent data-driven feedback loop. Its effectiveness is demonstrated via application to vibration compensation of a 3D printer with unmodeled linear and nonlinear dynamics. Thanks to the proposed hybrid FBF controller, the tracking accuracy of the 3D printer is significantly improved in experiments involving high-speed printing, compared to a standard FBF controller that does not incorporate a data-driven model. Furthermore, the ability of the hybrid FBF controller to detect and, hence, potentially avoid impending instability is demonstrated offline using data collected online from experiments.