Koopman invariant subspaces and finite linear representations of nonlinear dynamical systems for control

Abstract

In this work, we explore finite-dimensional linear representations of nonlinear dynamical systems by restricting the Koopman operator to an invariant subspace. The Koopman operator is an infinite-dimensional linear operator that evolves observable functions of the state-space of a dynamical system [Koopman 1931, PNAS]. Dominant terms in the Koopman expansion are typically computed using dynamic mode decomposition (DMD). DMD uses linear measurements of the state variables, and it has recently been shown that this may be too restrictive for nonlinear systems [Williams et al. 2015, JNLS]. Choosing nonlinear observable functions to form an invariant subspace where it is possible to obtain linear models, especially those that are useful for control, is an open challenge. Here, we investigate the choice of observable functions for Koopman analysis that enable the use of optimal linear control techniques on nonlinear problems. First, to include a cost on the state of the system, as in linear quadratic regulator (LQR) control, it is helpful to include these states in the observable subspace, as in DMD. However, we find that this is only possible when there is a single isolated fixed point, as systems with multiple fixed points or more complicated attractors are not globally topologically conjugate to a finite-dimensional linear system, and cannot be represented by a finite-dimensional linear Koopman subspace that includes the state. We then present a data-driven strategy to identify relevant observable functions for Koopman analysis using a new algorithm to determine terms in a dynamical system by sparse regression of the data in a nonlinear function space [Brunton et al. 2015, arxiv]; we show how this algorithm is related to DMD. Finally, we demonstrate how to design optimal control laws for nonlinear systems using techniques from linear optimal control on Koopman invariant subspaces.

20 pages, 5 figures, 2 codes

References in corpus (10)

• Discovering governing equations from data: Sparse identification of nonlinear dynamical systems
• The James Webb Space Telescope
• Applied Koopmanism
• Characterizing and correcting for the effect of sensor noise in the dynamic mode decomposition
• Dynamic Mode Decomposition for Large and Streaming Datasets
• Compressed Dynamic Mode Decomposition for Background Modeling
• Geometry of the ergodic quotient reveals coherent structures in flows
• Data Fusion via Intrinsic Dynamic Variables: An Application of Data-Driven Koopman Spectral Analysis
• Dynamic Mode Decomposition for Real-Time Background/Foreground Separation in Video
• Multi-Resolution Dynamic Mode Decomposition

Cited by in corpus (123)

• Deep learning for universal linear embeddings of nonlinear dynamics
• Linear predictors for nonlinear dynamical systems: Koopman operator meets model predictive control
• Chaos as an Intermittently Forced Linear System
• Data-Driven Sparse Sensor Placement for Reconstruction
• Ergodic theory, Dynamic Mode Decomposition and Computation of Spectral Properties of the Koopman operator
• Extended dynamic mode decomposition with dictionary learning: a data-driven adaptive spectral decomposition of the Koopman operator
• On Convergence of Extended Dynamic Mode Decomposition to the Koopman Operator
• Data-driven approximation of the Koopman generator: Model reduction, system identification, and control
• Sparse reduced-order modeling : Sensor-based dynamics to full-state estimation
• Discovery of Nonlinear Multiscale Systems: Sampling Strategies and Embeddings
• Data-Driven Modeling and Prediction of Non-Linearizable Dynamics via Spectral Submanifolds
• Multi-Scale Proper Orthogonal Decomposition of Complex Fluid Flows
• Physics-Informed Probabilistic Learning of Linear Embeddings of Non-linear Dynamics With Guaranteed Stability
• Eigendecompositions of Transfer Operators in Reproducing Kernel Hilbert Spaces
• Subspace dynamic mode decomposition for stochastic Koopman analysis
• Data-Driven Model Predictive Control using Interpolated Koopman Generators
• Residual Dynamic Mode Decomposition: Robust and verified Koopmanism
• Active Learning in Robotics: A Review of Control Principles
• Lagrangian Dynamic Mode Decomposition for Construction of Reduced-Order Models of Advection-Dominated Phenomena
• Neural-inspired sensors enable sparse, efficient classification of spatiotemporal data
• Sparse Identification of Nonlinear Dynamics with Control (SINDYc)
• Dynamic Feature-based Deep Reinforcement Learning for Flow Control of Circular Cylinder with Sparse Surface Pressure Sensing
• Physics-Informed Deep Learning For Traffic State Estimation: A Survey and the Outlook
• Sparsity-promoting algorithms for the discovery of informative Koopman invariant subspaces
• Koopman mode expansions between simple invariant solutions
• Data-driven Nonlinear Model Reduction to Spectral Submanifolds in Mechanical Systems
• Learning Compositional Koopman Operators for Model-Based Control
• From Fourier to Koopman: Spectral Methods for Long-term Time Series Prediction
• Deep Learning Models for Global Coordinate Transformations that Linearize PDEs
• Koopman-based feedback design with stability guarantees
• Data-Driven Participation Factors for Nonlinear Systems Based on Koopman Mode Decomposition
• Variational approach for learning Markov processes from time series data
• Searching turbulence for periodic orbits with dynamic mode decomposition
• Nonlinear Model Reduction to Fractional and Mixed-Mode Spectral Submanifolds
• Robust data-driven control for nonlinear systems using the Koopman operator
• Networked oscillator based modeling and control of unsteady wakes
• Learning Latent Dynamics for Partially-Observed Chaotic Systems
• Koopman analysis of Burgers equation
• Structured Time-Delay Models for Dynamical Systems with Connections to Frenet-Serret Frame
• Incorporating physical constraints in a deep probabilistic machine learning framework for coarse-graining dynamical systems
• Reduced-order Koopman modeling and predictive control of nonlinear processes
• Vehicular Applications of Koopman Operator Theory -- A Survey
• On analytical construction of observable functions in extended dynamic mode decomposition for nonlinear estimation and prediction
• Identification of MIMO Wiener-type Koopman Models for Data-Driven Model Reduction using Deep Learning
• Amidst data-driven model reduction and control
• Data-driven Control of Agent-based Models: an Equation/Variable-free Machine Learning Approach
• The Linear-Time-Invariance Notion of the Koopman Analysis-Part 1: The Architecture, Practical Rendering on the Prism Wake, and Fluid-Structure Association
• Approximate Analytical Solution to the Zonal Harmonics Problem Using Koopman Operator Theory
• Deep learning delay coordinate dynamics for chaotic attractors from partial observable data
• An overview of Koopman-based control: From error bounds to closed-loop guarantees
• Dynamical System Identification, Model Selection and Model Uncertainty Quantification by Bayesian Inference
• Anti-circulant dynamic mode decomposition with sparsity-promoting for highway traffic dynamics analysis
• Physics-informed machine learning for sensor fault detection with flight test data
• Nonequilibrium Statistical Mechanics and Optimal Prediction of Partially-Observed Complex Systems
• Global, Unified Representation of Heterogenous Robot Dynamics Using Composition Operators: A Koopman Direct Encoding Method
• Data-driven modelling of brain activity using neural networks, Diffusion Maps, and the Koopman operator
• Feedback control of nonlinear PDEs using data-efficient reduced order models based on the Koopman operator
• Auxiliary Functions as Koopman Observables: Data-Driven Analysis of Dynamical Systems via Polynomial Optimization
• Deep Identification of Nonlinear Systems in Koopman Form
• Controlling nonlinear PDEs using low-dimensional bilinear approximations obtained from data
• Backwards theory supports modelling via invariant manifolds for non-autonomous dynamical systems
• Koopman Representations of Dynamic Systems with Control
• Bounded nonlinear forecasts of partially observed geophysical systems with physics-constrained deep learning
• Data-Driven Optimal Control of Tethered Space Robot Deployment with Learning Based Koopman Operator
• Randomized Dynamic Mode Decomposition for Non-Intrusive Reduced Order Modelling
• Data-Driven Inference of High-Accuracy Isostable-Based Dynamical Models in Response to External Inputs
• Koopman Operator Theory for Nonlinear Dynamic Modeling using Dynamic Mode Decomposition
• Predictive Accuracy of Dynamic Mode Decomposition
• SafEDMD: A Koopman-based data-driven controller design framework for nonlinear dynamical systems
• Memory-Efficient Learning of Stable Linear Dynamical Systems for Prediction and Control
• Data-Driven Models for Control Engineering Applications Using the Koopman Operator
• Koopman Operator Dynamical Models: Learning, Analysis and Control
• Data-Driven Koopman Controller Synthesis Based on the Extended ${\mathcal{H}}_2$ Norm Characterization
• Dynamical System Segmentation for Information Measures in Motion
• Dynamic mode decomposition for interconnected control systems
• On the sample complexity of stabilizing linear dynamical systems from data
• Data-driven MPC with terminal conditions in the Koopman framework
• Generic Properties of Koopman Eigenfunctions for Stable Fixed Points and Periodic Orbits
• Entropy non-conservation and boundary conditions for Hamiltonian dynamical systems
• Koopman analysis in oscillator synchronization
• Markov Chain Monte Carlo for Koopman-based Optimal Control: Technical Report
• Modal Analysis of Fluid Flows: An Overview
• Koopman-based control of nonlinear systems with closed-loop guarantees
• Learning Koopman Representations for Hybrid Systems
• Koopman-based control using sum-of-squares optimization: Improved stability guarantees and data efficiency
• Data-Driven Batch Localization and SLAM Using Koopman Linearization
• Estimating Eigenenergies from Quantum Dynamics: A Unified Noise-Resilient Measurement-Driven Approach
• Kernel Autocovariance Operators of Stationary Processes: Estimation and Convergence
• When is Particle Filtering Efficient for Planning in Partially Observed Linear Dynamical Systems?
• Learning Data-Driven PCHD Models for Control Engineering Applications
• Challenges in Dynamic Mode Decomposition
• The Effect of Sensor Fusion on Data-Driven Learning of Koopman Operators
• Assessment of End-to-End and Sequential Data-driven Learning of Fluid Flows
• Dynamic Mode Decomposition for Construction of Reduced-Order Models of Hyperbolic Problems with Shocks
• Kernel Embedding based Variational Approach for Low-dimensional Approximation of Dynamical Systems
• Painting the Phase Space of Dissipative Systems with Lagrangian Descriptors
• An information-geometric approach to feature extraction and moment reconstruction in dynamical systems
• System stabilization with policy optimization on unstable latent manifolds
• Participation Factors for Nonlinear Autonomous Dynamical Systems in the Koopman Operator Framework
• $Q$-functions, synchronization, and Arnold tongues for coupled stochastic oscillators
• Interpretable and flexible non-intrusive reduced-order models using reproducing kernel Hilbert spaces
• Robust Control Design and Analysis Based on Lifting Linearization of Nonlinear Systems Under Uncertain Initial Conditions
• Diffusion Maps meet Nyström
• Data-driven Model Predictive Control Method for DFIG-based Wind Farm to Provide Primary Frequency Regulation Service
• Enhancement for Robustness of Koopman Operator-based Data-driven Mobile Robotic Systems
• A Data-driven Hierarchical Control Structure for Systems with Uncertainty
• Sparsity enabled cluster reduced-order models for control
• Koopman operator approach for computing structure of solutions and Observability of non-linear finite state system
• Predicting sparse circle maps from their dynamics
• Koopman Operator Based Modeling for Quadrotor Control on $SE(3)$
• Kalman Filter Aided Federated Koopman Learning
• A Methodology for Projection-Based Model Reduction with Black-Box High-Fidelity Models
• Data-Driven Contact-Aware Control Method for Real-Time Deformable Tool Manipulation: A Case Study in the Environmental Swabbing
• Inertial-Based LQG Control: A New Look at Inverted Pendulum Stabilization
• Efficient pseudometrics for data-driven comparisons of nonlinear dynamical systems
• Improving Estimation of the Koopman Operator with Kolmogorov-Smirnov Indicator Functions
• On Koopman Operator for Burgers' Equation
• Reduced-Order Surrogates for Forced Flexible Mesh Coastal-Ocean Models
• A purely data-driven framework for prediction, optimization, and control of networked processes: application to networked SIS epidemic model
• Comparing Dynamical Models Through Diffeomorphic Vector Field Alignment
• Paradigm Shift Through the Integration of Physical Methodology and Data Science
• Far-Field Minimum-Fuel Spacecraft Rendezvous using Koopman Operator and ${\ell }_2/{\ell }_1$ Optimization
• Physical discovery in representation learning via conditioning on prior knowledge: applications for ferroelectric domain dynamics