CaltechAUTHORS
  A Caltech Library Service

Neural Stochastic Contraction Metrics for Learning-based Control and Estimation

Tsukamoto, Hiroyasu and Chung, Soon-Jo and Slotine, Jean-Jacques E. (2021) Neural Stochastic Contraction Metrics for Learning-based Control and Estimation. IEEE Control Systems Letters, 5 (5). pp. 1825-1830. ISSN 2475-1456. https://resolver.caltech.edu/CaltechAUTHORS:20210113-163505450

[img] PDF - Accepted Version
See Usage Policy.

946Kb

Use this Persistent URL to link to this item: https://resolver.caltech.edu/CaltechAUTHORS:20210113-163505450

Abstract

We present Neural Stochastic Contraction Metrics (NSCM), a new design framework for provably-stable learning-based control and estimation for a class of stochastic nonlinear systems. It uses a spectrally-normalized deep neural network to construct a contraction metric and its differential Lyapunov function, sampled via simplified convex optimization in the stochastic setting. Spectral normalization constrains the state-derivatives of the metric to be Lipschitz continuous, thereby ensuring exponential boundedness of the mean squared distance of system trajectories under stochastic disturbances. The trained NSCM model allows autonomous systems to approximate optimal stable control and estimation policies in real-time, and outperforms existing nonlinear control and estimation techniques including the state-dependent Riccati equation, iterative LQR, EKF, and the deterministic NCM, as shown in simulation results.


Item Type:Article
Related URLs:
URLURL TypeDescription
https://doi.org/10.1109/lcsys.2020.3046529DOIArticle
https://arxiv.org/abs/2011.03168arXivDiscussion Paper
https://github.com/astrohiro/nscmRelated ItemCode
ORCID:
AuthorORCID
Chung, Soon-Jo0000-0002-6657-3907
Slotine, Jean-Jacques E.0000-0002-7161-7812
Additional Information:© 2020 IEEE. Manuscript received September 14, 2020; revised November 18, 2020; accepted December 7, 2020. Date of publication December 22, 2020; date of current version January 13, 2021. This work was supported in part by the Raytheon Company. This work was benefited from discussions with Nicholas Boffi and Quang-Cuong Pham. Code: https://github.com/astrohiro/nscm.
Group:GALCIT, Center for Autonomous Systems and Technologies (CAST)
Funders:
Funding AgencyGrant Number
Raytheon CompanyUNSPECIFIED
Subject Keywords:Machine learning, stochastic optimal control, observers for nonlinear systems
Issue or Number:5
Record Number:CaltechAUTHORS:20210113-163505450
Persistent URL:https://resolver.caltech.edu/CaltechAUTHORS:20210113-163505450
Official Citation:H. Tsukamoto, S. -J. Chung and J. -J. E. Slotine, "Neural Stochastic Contraction Metrics for Learning-Based Control and Estimation," in IEEE Control Systems Letters, vol. 5, no. 5, pp. 1825-1830, Nov. 2021, doi: 10.1109/LCSYS.2020.3046529
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:107468
Collection:CaltechAUTHORS
Deposited By: George Porter
Deposited On:14 Jan 2021 17:55
Last Modified:14 Jan 2021 17:55

Repository Staff Only: item control page