TU Darmstadt / ULB / TUprints

Optimal Control and Inverse Optimal Control by Distribution Matching

Arenz, Oleg ; Abdulsamad, Hany ; Neumann, Gerhard (2022):
Optimal Control and Inverse Optimal Control by Distribution Matching. (Postprint)
In: 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp. 4046-4053,
Darmstadt, IEEE, 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Daejeon, Korea, 09.-14.10.2016, e-ISSN 2153-0866, ISBN 978-1-5090-3762-9,
DOI: 10.26083/tuprints-00022929,
[Conference or Workshop Item]

[img] Text
Optimal_control_and_inverse_optimal_control_by_distribution_matching.pdf
Copyright Information: In Copyright.

Download (923kB)
Item Type: Conference or Workshop Item
Origin: Secondary publication service
Status: Postprint
Title: Optimal Control and Inverse Optimal Control by Distribution Matching
Language: English
Abstract:

Optimal control is a powerful approach to achieve optimal behavior. However, it typically requires a manual specification of a cost function which often contains several objectives, such as reaching goal positions at different time steps or energy efficiency. Manually trading-off these objectives is often difficult and requires a high engineering effort. In this paper, we present a new approach to specify optimal behavior. We directly specify the desired behavior by a distribution over future states or features of the states. For example, the experimenter could choose to reach certain mean positions with given accuracy/variance at specified time steps. Our approach also unifies optimal control and inverse optimal control in one framework. Given a desired state distribution, we estimate a cost function such that the optimal controller matches the desired distribution. If the desired distribution is estimated from expert demonstrations, our approach performs inverse optimal control. We evaluate our approach on several optimal and inverse optimal control tasks on non-linear systems using incremental linearizations similar to differential dynamic programming approaches.

Book Title: 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)
Place of Publication: Darmstadt
Publisher: IEEE
Collation: 14 ungezählte Seiten
Uncontrolled Keywords: Optimal control, Entropy, Heuristic algorithms, Trajectory, Cost function, Learning (artificial intelligence)
Classification DDC: 000 Allgemeines, Informatik, Informationswissenschaft > 004 Informatik
Divisions: 20 Department of Computer Science > Intelligent Autonomous Systems
Event Title: 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)
Event Location: Daejeon, Korea
Event Dates: 09.-14.10.2016
Date Deposited: 25 Nov 2022 12:51
Last Modified: 13 Dec 2022 08:09
DOI: 10.26083/tuprints-00022929
Corresponding Links:
URN: urn:nbn:de:tuda-tuprints-229290
URI: https://tuprints.ulb.tu-darmstadt.de/id/eprint/22929
PPN:
Export:
Actions (login required)
View Item View Item