Koopman NMPC: Koopman-based Learning and Nonlinear Model Predictive Control of Control-affine Systems

Koopman-based learning methods can potentially be practical and powerful tools for dynamical robotic systems. However, common methods to construct Koopman representations seek to learn lifted linear models that cannot capture nonlinear actuation effects inherent in many robotic systems. This paper p...

Full description

Saved in:
Bibliographic Details
Published in2021 IEEE International Conference on Robotics and Automation (ICRA) pp. 7350 - 7356
Main Authors Folkestad, Carl, Burdick, Joel W.
Format Conference Proceeding
LanguageEnglish
Published IEEE 30.05.2021
Subjects
Computational modeling
Dictionaries
Learning systems
Predictive models
System dynamics
Transforms
Online AccessGet full text

Cover

More Information
Summary:Koopman-based learning methods can potentially be practical and powerful tools for dynamical robotic systems. However, common methods to construct Koopman representations seek to learn lifted linear models that cannot capture nonlinear actuation effects inherent in many robotic systems. This paper presents a learning and control methodology that is a first step towards overcoming this limitation. Using the Koopman canonical transform, control-affine dynamics can be expressed by a lifted bilinear model. The learned model is used for nonlinear model predictive control (NMPC) design where the bilinear structure can be exploited to improve computational efficiency. The benefits for control-affine dynamics compared to existing Koopman-based methods are highlighted through an example of a simulated planar quadrotor. Prediction error is greatly reduced and closed loop performance similar to NMPC with full model knowledge is achieved.
ISSN:2577-087X
DOI:10.1109/ICRA48506.2021.9562002