Navigation überspringen
Universitätsbibliothek Heidelberg
Status: Bibliographieeintrag

Verfügbarkeit
Standort: ---
Exemplare: ---
heiBIB
 Online-Ressource
Verfasst von:Koryakovskiy, Ivan [VerfasserIn]   i
 Kudruss, Manuel [VerfasserIn]   i
 Kirches, Christian [VerfasserIn]   i
 Mombaur, Katja [VerfasserIn]   i
 Schlöder, Johannes P. [VerfasserIn]   i
Titel:Benchmarking model-free and model-based optimal control
Verf.angabe:Ivan Koryakovskiy, Manuel Kudruss, Robert Babuška, Wouter Caarls, Christian Kirches, Katja Mombaur, Johannes P. Schlöder, Heike Vallery
E-Jahr:2017
Jahr:2 March 2017
Umfang:10 S.
Fussnoten:Available online 2 March 2017 ; Gesehen am 25.10.2018
Titel Quelle:Enthalten in: Robotics and autonomous systems
Ort Quelle:Amsterdam [u.a.] : Elsevier, 1988
Jahr Quelle:2017
Band/Heft Quelle:92(2017), Seite 81-90
ISSN Quelle:1872-793X
Abstract:Model-free reinforcement learning and nonlinear model predictive control are two different approaches for controlling a dynamic system in an optimal way according to a prescribed cost function. Reinforcement learning acquires a control policy through exploratory interaction with the system, while nonlinear model predictive control exploits an explicitly given mathematical model of the system. In this article, we provide a comprehensive comparison of the performance of reinforcement learning and nonlinear model predictive control for an ideal system as well as for a system with parametric and structural uncertainties. The comparison is based on two different criteria, namely the similarity of trajectories and the resulting rewards. The evaluation of both methods is performed on a standard benchmark problem: a cart-pendulum swing-up and balance task. We first find suitable mathematical formulations and discuss the effect of the differences in the problem formulations. Then, we investigate the robustness of reinforcement learning and nonlinear model predictive control against uncertainties. The results demonstrate that nonlinear model predictive control has advantages over reinforcement learning if uncertainties can be eliminated through identification of the system parameters. Otherwise, there exists a break-even point after which model-free reinforcement learning performs better than nonlinear model predictive control with an inaccurate model. These findings suggest that benefits can be obtained by combining these methods for real systems being subject to such uncertainties. In the future, we plan to develop a hybrid controller and evaluate its performance on a real seven-degree-of-freedom walking robot.
DOI:doi:10.1016/j.robot.2017.02.006
URL:Bitte beachten Sie: Dies ist ein Bibliographieeintrag. Ein Volltextzugriff für Mitglieder der Universität besteht hier nur, falls für die entsprechende Zeitschrift/den entsprechenden Sammelband ein Abonnement besteht oder es sich um einen OpenAccess-Titel handelt.

Volltext ; Verlag: http://dx.doi.org/10.1016/j.robot.2017.02.006
 Volltext: http://www.sciencedirect.com/science/article/pii/S0921889016301592
 DOI: https://doi.org/10.1016/j.robot.2017.02.006
Datenträger:Online-Ressource
Sprache:eng
Sach-SW:Nonlinear model predictive control
 Optimal control
 Parametric uncertainties
 Reinforcement learning
 Structural uncertainties
K10plus-PPN:1582286280
Verknüpfungen:→ Zeitschrift

Permanenter Link auf diesen Titel (bookmarkfähig):  https://katalog.ub.uni-heidelberg.de/titel/68321198   QR-Code
zum Seitenanfang