Learning stabilizable nonlinear dynamics with contraction-based regularization
Open accessThe DOI matches its Crossref record; 5 of the 5 references it cites were found in a registry or an index.
Abstract
We propose a novel framework for learning stabilizable nonlinear dynamical systems for continuous control tasks in robotics. The key contribution is a control-theoretic regularizer for dynamics fitting rooted in the notion of stabilizability, a constraint which guarantees the existence of robust tracking controllers for arbitrary open-loop trajectories generated with the learned system. Leveraging tools from contraction theory and statistical learning in reproducing kernel Hilbert spaces, we formulate stabilizable dynamics learning as a functional optimization with a convex objective and bi-convex functional constraints. Under a mild structural assumption and relaxation of the functional constraints to sampling-based constraints, we derive the optimal solution with a modified representer theorem. Finally, we utilize random matrix feature approximations to reduce the dimensionality of the search parameters and formulate an iterative convex optimization algorithm that jointly fits the dynamics functions and searches for a certificate of stabilizability. We validate the proposed algorithm in simulation for a planar quadrotor, and on a quadrotor hardware testbed emulating planar dynamics. We verify, both in simulation and on hardware, significantly improved trajectory generation and tracking performance with the control-theoretic regularized model over models learned using traditional regression techniques, especially when learning from small supervised datasets. The results support the conjecture that the use of stabilizability constraints as a form of regularization can help prune the hypothesis space in a manner that is tailored to the downstream task of trajectory generation and feedback control. This produces models that are not only dramatically better conditioned, but also data efficient.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Learning Stable Nonlinear Dynamical Systems With Gaussian Mixture Models
Khansari-Zadeh, S. M., & Billard, A. (2011) · IEEE Transactions on Robotics · doi:10.1109/tro.2011.2159412 · Cited by 771 · cited 2×
“In this line of work, one assumes perfect knowledge and invertibility of the robot’s controlled dynamics to solve for the input that realizes this desirable closed-loop motion (Lemme et al., 2014; Khansari-Zadeh and Khatib, 2017; Ravichandar et al., 2017; Khansari-Zadeh and Billard, 2011; Medina and Billard, 2017).”
BackgroundDeep reinforcement learning in a handful of trials using probabilistic dynamics models
Chua, K., Calandra, R., McAllister, R., & Levine, S. (2018) · arXiv · doi:10.48550/arxiv.1805.12114 · Cited by 525 · cited 6×
“(Chua et al., 2018) leverage an ensemble of probabilistic models to capture both epistemic (i.”
BackgroundMethodInfluentialDynamics and control of a class of underactuated mechanical systems
Reyhanoglu, M., van der Schaft, A., Mcclamroch, N. H., & Kolmanovsky, I. (1999) · IEEE Transactions on Automatic Control · doi:10.1109/9.788533 · Cited by 501 · cited 3×
“By rearranging the configuration vector (Spong, 1998; Olfati-Saber, 2001; Reyhanoglu et al., 1999), one can partition q as (qu, qa) where qu ∈ Rnq−m represents the unactuated degrees of freedom and qa ∈ Rm represents the actuated degrees of freedom.”
BackgroundSpong (1998) · doi:10.1007/bfb0015081 · Cited by 437 · cited 1×
“By rearranging the configuration vector (Spong, 1998; Olfati-Saber, 2001; Reyhanoglu et al., 1999), one can partition q as ( q u , q a ) where q u ∈ R n q − m represents the unactuated degrees of freedom and q a ∈ R m represents the actuated degrees of freedom.”
BackgroundImproving Multi-Step Prediction of Learned Time Series Models
Venkatraman, A., Hebert, M., & Bagnell, J. (2015) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v29i1.9590 · Cited by 198
Cited by
Dawson, C., Gao, S., & Fan, C. (2023) · IEEE Transactions on Robotics · doi:10.1109/tro.2022.3232542 · Cited by 269
Contraction theory for nonlinear stability analysis and learning-based control: A tutorial overview
Tsukamoto, H., Chung, S.-J., & Slotine, J.-J. E. (2021) · Annual Reviews in Control · doi:10.1016/j.arcontrol.2021.10.001 · Cited by 145
Self-citationNeural Operators for Bypassing Gain and Control Computations in PDE Backstepping
Bhan, L., Shi, Y., & Krstic, M. (2024) · IEEE Transactions on Automatic Control · doi:10.1109/tac.2023.3347499 · Cited by 58
“The main focus has been on learning the system dynamics and providing closed-loopguaranteesin finite-time forbothlinearsystems[15], [23], [29], [42], [78] (and references within); and nonlinear systems [5], [35], [43], [72].”
BackgroundRobot Model Identification and Learning: A Modern Perspective
Lee, T., Kwon, J., Wensing, P. M., & Park, F. C. (2024) · Annual Review of Control, Robotics, and Autonomous Systems · doi:10.1146/annurev-control-061523-102310 · Cited by 34
Metadata
- Authors
- Singh, S., Richards, S. M., Sindhwani, V., Slotine, J.-J. E., & Pavone, M.
- Venue
- The International Journal of Robotics Research
- Publisher
- SAGE Publications
- Published
- 2020-08-30
- Volume
- 40
- Issue
- 10-11
- Pages
- 1123-1150
- ISSN
- 0278-3649 (ISSN-L)
- 1741-3176
- OpenAlex
- W2965195404
- Cited by
- 64
- References
- 5
- FWCI
- 3.79
- Type
- journal article
- Language
- en
- Open access
- green
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.