Guaranteed Trajectory Tracking under Learned Dynamics with Contraction Metrics and Disturbance Estimation

Zhao, Pan; Guo, Ziyao; Cheng, Yikun; Gahlawat, Aditya; Kang, Hyungsoo; Hovakimyan, Naira

doi:10.3390/robotics13070099

Open AccessArticle

Guaranteed Trajectory Tracking under Learned Dynamics with Contraction Metrics and Disturbance Estimation

by

Pan Zhao

^1,*

,

Ziyao Guo

²

,

Yikun Cheng

²

,

Aditya Gahlawat

²

,

Hyungsoo Kang

² and

Naira Hovakimyan

²

¹

Department of Aerospace Engineering and Mechanics, University of Alabama, Tuscaloosa, AL 35487, USA

²

Department of Mechanical Science and Engineering, University of Illinois at Urbana-Champaign, Urbana, IL 61801, USA

^*

Author to whom correspondence should be addressed.

Robotics 2024, 13(7), 99; https://doi.org/10.3390/robotics13070099

Submission received: 13 May 2024 / Revised: 19 June 2024 / Accepted: 25 June 2024 / Published: 30 June 2024

(This article belongs to the Special Issue Decision-Making and Control under Uncertainties for Robotic and Autonomous Systems)

Download

Browse Figures

Versions Notes

Abstract

This paper presents a contraction-based learning control architecture that allows for using model learning tools to learn matched model uncertainties while guaranteeing trajectory tracking performance during the learning transients. The architecture relies on a disturbance estimator to estimate the pointwise value of the uncertainty, i.e., the discrepancy between a nominal model and the true dynamics, with pre-computable estimation error bounds, and a robust Riemannian energy condition for computing the control signal. Under certain conditions, the controller guarantees exponential trajectory convergence during the learning transients, while learning can improve robustness and facilitate better trajectory planning. Simulation results validate the efficacy of the proposed control architecture.

Keywords: robust control; decision-making under uncertainty; machine learning for control; robot safety

Share and Cite

MDPI and ACS Style

Zhao, P.; Guo, Z.; Cheng, Y.; Gahlawat, A.; Kang, H.; Hovakimyan, N. Guaranteed Trajectory Tracking under Learned Dynamics with Contraction Metrics and Disturbance Estimation. Robotics 2024, 13, 99. https://doi.org/10.3390/robotics13070099

AMA Style

Zhao P, Guo Z, Cheng Y, Gahlawat A, Kang H, Hovakimyan N. Guaranteed Trajectory Tracking under Learned Dynamics with Contraction Metrics and Disturbance Estimation. Robotics. 2024; 13(7):99. https://doi.org/10.3390/robotics13070099

Chicago/Turabian Style

Zhao, Pan, Ziyao Guo, Yikun Cheng, Aditya Gahlawat, Hyungsoo Kang, and Naira Hovakimyan. 2024. "Guaranteed Trajectory Tracking under Learned Dynamics with Contraction Metrics and Disturbance Estimation" Robotics 13, no. 7: 99. https://doi.org/10.3390/robotics13070099

APA Style

Zhao, P., Guo, Z., Cheng, Y., Gahlawat, A., Kang, H., & Hovakimyan, N. (2024). Guaranteed Trajectory Tracking under Learned Dynamics with Contraction Metrics and Disturbance Estimation. Robotics, 13(7), 99. https://doi.org/10.3390/robotics13070099

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Menu

Guaranteed Trajectory Tracking under Learned Dynamics with Contraction Metrics and Disturbance Estimation

Abstract

Share and Cite

Article Metrics

Article Access Statistics

Further Information

Guidelines

MDPI Initiatives

Follow MDPI