Cautious Model Predictive Control Using Gaussian Process Regression


Loading...

Date

2020-11

Publication Type

Journal Article

ETH Bibliography

yes

Citations

Altmetric

Data

Abstract

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.

Publication status

published

Editor

Book title

Volume

28 (6)

Pages / Article No.

1237 - 1259

Publisher

IEEE

Event

Edition / version

Methods

Software

Geographic location

Date collected

Date created

Subject

Model Predictive Control; Gaussian Processes; Learning-based Control; Model Learning; Autonomous Racing

Organisational unit

09563 - Zeilinger, Melanie / Zeilinger, Melanie check_circle

Notes

Funding

157601 - Safety and Performance for Human in the Loop Control (SNF)

Related publications and datasets