Integrated Pre-Processing for Bayesian Nonlinear System Identification with Gaussian Processes

TL;DR

This paper presents GP-FNARX, an automated Bayesian system identification method that combines data pre-processing with Gaussian processes to model nonlinear dynamics and quantify uncertainty, suitable for robotics and control.

Contribution

It introduces a fully automated, integrated approach for nonlinear system identification using Gaussian processes with pre-processing and hyper-parameter tuning.

Findings

Successfully models nonlinear systems with uncertainty quantification.

Automated procedure from raw data to identified model.

Low computational cost for practical applications.

Abstract

We introduce GP-FNARX: a new model for nonlinear system identification based on a nonlinear autoregressive exogenous model (NARX) with filtered regressors (F) where the nonlinear regression problem is tackled using sparse Gaussian processes (GP). We integrate data pre-processing with system identification into a fully automated procedure that goes from raw data to an identified model. Both pre-processing parameters and GP hyper-parameters are tuned by maximizing the marginal likelihood of the probabilistic model. We obtain a Bayesian model of the system's dynamics which is able to report its uncertainty in regions where the data is scarce. The automated approach, the modeling of uncertainty and its relatively low computational cost make of GP-FNARX a good candidate for applications in robotics and adaptive control.