Data-Driven Sparse Sensor Placement for Reconstruction

Abstract

Optimal sensor placement is a central challenge in the design, prediction, estimation, and control of high-dimensional systems. High-dimensional states can often leverage a latent low-dimensional representation, and this inherent compressibility enables sparse sensing. This article explores optimized sensor placement for signal reconstruction based on a tailored library of features extracted from training data. Sparse point sensors are discovered using the singular value decomposition and QR pivoting, which are two ubiquitous matrix computations that underpin modern linear dimensionality reduction. Sparse sensing in a tailored basis is contrasted with compressed sensing, a universal signal recovery method in which an unknown signal is reconstructed via a sparse representation in a universal basis. Although compressed sensing can recover a wider class of signals, we demonstrate the benefits of exploiting known patterns in data with optimized sensing. In particular, drastic reductions in the required number of sensors and improved reconstruction are observed in examples ranging from facial images to fluid vorticity fields. Principled sensor placement may be critically enabling when sensors are costly and provides faster state estimation for low-latency, high-bandwidth control. MATLAB code is provided for all examples.

34 pages, 17 figures

References in corpus (5)

• Applied Koopmanism
• Model selection for dynamical systems via sparse regression and information criteria
• Network Structure of Two-Dimensional Decaying Isotropic Turbulence
• Network-theoretic approach to sparsified discrete vortex dynamics
• Data Fusion via Intrinsic Dynamic Variables: An Application of Data-Driven Koopman Spectral Analysis

Cited by in corpus (67)

• Machine Learning for Fluid Mechanics
• Uncovering near-wall blood flow from sparse data with physics-informed neural networks
• Shallow Neural Networks for Fluid Flow Reconstruction with Limited Sensors
• Global field reconstruction from sparse sensors with Voronoi tessellation-assisted deep learning
• Super-resolution and denoising of fluid flow using physics-informed convolutional neural networks without high-resolution labels
• Robust flow field reconstruction from limited measurements via sparse representation
• Super-Resolution Analysis via Machine Learning: A Survey for Fluid Flows
• Sparse Principal Component Analysis via Variable Projection
• Randomized Dynamic Mode Decomposition
• Generalization techniques of neural networks for fluid flow estimation
• Determinant-based Fast Greedy Sensor Selection Algorithm
• Deep Model Predictive Control with Online Learning for Complex Physical Systems
• Predicting shim gaps in aircraft assembly with machine learning and sparse sensing
• Data-driven cardiovascular flow modeling: examples and opportunities
• Neural-inspired sensors enable sparse, efficient classification of spatiotemporal data
• Fast Greedy Optimization of Sensor Selection in Measurement with Correlated Noise
• Leveraging reduced-order models for state estimation using deep learning
• Optimized Sampling for Multiscale Dynamics
• Data-driven modeling of the chaotic thermal convection in an annular thermosyphon
• A new approach for determining optimal placement of PM2.5 air quality sensors: case study for the contiguous United States
• Data-Driven Approach for Noise Reduction in Pressure-Sensitive Paint Data Based on Modal Expansion and Time-Series Data at Optimally Placed Points
• Effect of Objective Function on Data-Driven Greedy Sparse Sensor Optimization
• Deep-learning assisted reduced order model for high-dimensional flow prediction from sparse data
• Data-driven sparse sensor placement based on A-optimal design of experiment with ADMM
• Data-driven Vector-measurement-sensor Selection based on Greedy Algorithm
• Data-Driven Optimal Sensor Placement for High-Dimensional System Using Annealing Machine
• Interplay of Sensor Quantity, Placement and System Dimensionality on Energy Sparse Reconstruction of Fluid Flows
• Randomized Subspace Newton Convex Method Applied to Data-Driven Sensor Selection Problem
• A Machine Learning Surrogate Modeling Benchmark for Temperature Field Reconstruction of Heat-Source Systems
• A DeepONet multi-fidelity approach for residual learning in reduced order modeling
• Data-Driven Sensor Selection Method Based on Proximal Optimization for High-Dimensional Data With Correlated Measurement Noise
• Semi Conditional Variational Auto-Encoder for Flow Reconstruction and Uncertainty Quantification from Limited Observations
• Reinforcement-learning-based control of convectively-unstable flows
• Nondominated-Solution-based Multi-objective Greedy Sensor Selection for Optimal Design of Experiments
• Equation-free surrogate modeling of geophysical flows at the intersection of machine learning and data assimilation
• Data-Driven Selection of Actuators for Optimal Control of Airfoil Separation
• Seismic Wavefield Reconstruction based on Compressed Sensing using Data-Driven Reduced-Order Model
• Optimization of Sparse Sensor Placement for Estimation of Wind Direction and Surface Pressure Distribution Using Time-Averaged Pressure-Sensitive Paint Data on Automobile Model
• Proof-of-concept Study of Sparse Processing Particle Image Velocimetry for Real Time Flow Observation
• Adaptive sparse interpolation for accelerating nonlinear stochastic reduced-order modeling with time-dependent bases
• Robust training approach of neural networks for fluid flow state estimations
• Tensor-based flow reconstruction from optimally located sensor measurements
• Optimal sensor placement for reconstructing wind pressure field around buildings using compressed sensing
• Randomized Group-Greedy Method for Large-Scale Sensor Selection Problems
• Towards real-time reconstruction of velocity fluctuations in turbulent channel flow
• Sensitivity-enhanced generalized polynomial chaos for efficient uncertainty quantification
• Inadequacy of Linear Methods for Minimal Sensor Placement and Feature Selection in Nonlinear Systems; a New Approach Using Secants
• A data-based, reduced-order, dynamic estimator for reconstruction of non-linear flows exhibiting limit-cycle oscillations
• Optical neuromorphic computing via temporal up-sampling and trainable encoding on a telecom device platform
• Robust State Estimation from Partial Out-Core Measurements with Shallow Recurrent Decoder for Nuclear Reactors
• Data-driven prediction of multistable systems from sparse measurements
• Real-time forecasting of chaotic dynamics from sparse data and autoencoders
• Observation Site Selection for Physical Model Parameter Estimation toward Process-Driven Seismic Wavefield Reconstruction
• Model-based multi-sensor fusion for reconstructing wall-bounded turbulence
• Some effects of limited wall-sensor availability on flow estimation with 3D-GANs
• Polynomial Ridge Flowfield Estimation
• Fast Data-driven Greedy Sensor Selection for Ridge Regression
• Compact representation of transonic airfoil buffet flows with observable-augmented machine learning
• A fast food-freezing temperature estimation framework using optimally located sensors
• A simulation-free seismic survey design by maximizing the spectral gap
• Triple decomposition and sparse representation for noisy pressure-sensitive paint data
• Computationally and Memory-Efficient Robust Predictive Analytics Using Big Data
• Data-driven prediction of reversal of large-scale circulation in turbulent convection
• Time-Vertex Machine Learning for Optimal Sensor Placement in Temporal Graph Signals: Applications in Structural Health Monitoring
• An efficient offline sensor placement method for flow estimation
• Data-driven Sensor Placement for Predictive Applications: A Correlation-Assisted Attribution Framework (CAAF)
• 3D seismic survey design by maximizing the spectral gap