An LMI framework for contraction-based nonlinear control design by derivatives of Gaussian process regression

DOI10.1016/J.AUTOMATICA.2023.110928zbMATH Open1520.93214arXiv2301.08398MaRDI QIDQ6164037FDOQ6164037

Authors: Yu Kawano, Kenji Kashima 

Publication date: 30 June 2023

Published in: Automatica (Search for Journal in Brave)

Abstract: Contraction theory formulates the analysis of nonlinear systems in terms of Jacobian matrices. Although this provides the potential to develop a linear matrix inequality (LMI) framework for nonlinear control design, conditions are imposed not on controllers but on their partial derivatives, which makes control design challenging. In this paper, we illustrate this so-called integrability problem can be solved by a non-standard use of Gaussian process regression (GPR) for parameterizing controllers and then establish an LMI framework of contraction-based control design for nonlinear discrete-time systems, as an easy-to-implement tool. Later on, we consider the case where the drift vector fields are unknown and employ GPR for functional fitting as its standard use. GPR describes learning errors in terms of probability, and thus we further discuss how to incorporate stochastic learning errors into the proposed LMI framework.

Full work available at URL: https://arxiv.org/abs/2301.08398

Recommendations

• LMI conditions for contraction, integral action, and output feedback stabilization for a class of nonlinear systems
• Contraction analysis of non-linear distributed systems
• Infinite gain margin, contraction and optimality: an LMI-based design
• Distributed nonlinear model predictive control based on contraction theory
• A contraction property for state feedback design of linear discrete-time systems

zbMATH Keywords

Gaussian process regressiondiscrete-time systemsnonlinear systemsstochastic systemscontraction analysis

Mathematics Subject Classification ID

Nonlinear systems in control theory (93C10) Discrete-time control/observation systems (93C55) Stochastic systems in control theory (general) (93E03)

Cites Work

• Gaussian processes for machine learning.
• Linear Matrix Inequalities in System and Control Theory
• Pattern recognition and machine learning.
• Deep learning
• Transverse Exponential Stability and Applications
• A Differential Lyapunov Framework for Contraction Analysis
• Kernel methods in system identification, machine learning and function estimation: a survey
• On contraction analysis for non-linear systems
• Contraction Analysis of Monotone Systems via Separable Functions
• Differentially Positive Systems
• On using least-squares updates without regressor filtering in identification and adaptive control of nonlinear systems
• Controller Reduction for Nonlinear Systems by Generalized Differential Balancing
• Differential Dissipativity Theory for Dominance Analysis
• Stable Gaussian process based tracking control of Euler-Lagrange systems
• Online Bayesian inference and learning of Gaussian-process state-space models
• Control Contraction Metrics: Convex and Intrinsic Criteria for Nonlinear Feedback Design
• Feedback Linearization Based on Gaussian Processes With Event-Triggered Online Learning
• Convergence Properties for Discrete-Time Nonlinear Systems
• Contraction analysis of virtually positive systems
• Krasovskii and Shifted Passivity-Based Control
• Nonlinear Eigenvalue Approach to Differential Riccati Equations for Contraction Analysis
• Model Reduction by Differential Balancing Based on Nonlinear Hankel Operators
• Sufficient Conditions for Global Integral Action via Incremental Forwarding for Input-Affine Nonlinear Systems

This page was built for publication: An LMI framework for contraction-based nonlinear control design by derivatives of Gaussian process regression

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6164037)