Cautious Model Predictive Control Using Gaussian Process Regression

doi:10.1109/TCST.2019.2949757

Cautious Model Predictive Control Using Gaussian Process Regression

2020-11-01

378

OA

AI

DOI：10.1109/TCST.2019.2949757

Original

Save

View PDF

Abstract

En 中文

Gaussian process (GP) regression has been widely used in supervised machine learning due to its flexibility and inherent ability to describe uncertainty in function estimation. In the context of control, it is seeing increasing use for modeling of nonlinear dynamical systems from data, as it allows the direct assessment of residual model uncertainty. We present a model predictive control (MPC) approach that integrates a nominal system with an additive nonlinear part of the dynamics modeled as a GP. We describe a principled way of formulating the chance-constrained MPC problem, which takes into account residual uncertainties provided by the GP model to enable cautious control. Using additional approximations for efficient computation, we finally demonstrate the approach in a simulation example, as well as in a hardware implementation for autonomous racing of remote-controlled race cars with fast sampling times of 20 ms, highlighting improvements with regard to both performance and safety over a nominal controller.

Keywords:

Predictive control

Data models

Computational modeling

Kernel

Gaussian processes

Uncertainty

Predictive models

Autonomous racing

Gaussian processes (GPs)

learning-based control

model learning

model predictive control (MPC)

Cautious Model Predictive Control Using Gaussian Process Regression

Abstract

AI Summary

Journal

Organization