Data-Driven Sparse Sensor Placement for Reconstruction DEMONSTRATING THE BENEFITS OF EXPLOITING KNOWN PATTERNS

被引:310
作者
Manohar, Krithika [1 ]
Brunton, Bingni W. [2 ]
Kutz, J. Nathan [2 ]
Brunton, Steven L. [2 ]
机构
[1] Univ Washington, Appl Math, Seattle, WA 98195 USA
[2] Univ Washington, eSci Inst, Seattle, WA 98195 USA
来源
IEEE CONTROL SYSTEMS MAGAZINE | 2018年 / 38卷 / 03期
关键词
DYNAMIC-MODE DECOMPOSITION; EMPIRICAL INTERPOLATION METHOD; IMMERSED BOUNDARY METHOD; SIGNAL RECOVERY; FACE RECOGNITION; SPECTRAL PROPERTIES; REDUCTION; SYSTEMS; IDENTIFICATION; APPROXIMATION;
D O I
10.1109/MCS.2018.2810460
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
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 on a tailored basis is contrasted with compressed sensing, a universal signal recovery method in which an unknown signal is reconstructed via a sparse representation on 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.
引用
收藏
页码:63 / 86
页数:24
相关论文
共 148 条
[1]  
Akaike H, 1973, Selected Papers of Hirotugu Akaike, P199, DOI DOI 10.1007/978-1-4612-1694-0_15
[2]   Nonlinear model order reduction based on local reduced-order bases [J].
Amsallem, David ;
Zahr, Matthew J. ;
Farhat, Charbel .
INTERNATIONAL JOURNAL FOR NUMERICAL METHODS IN ENGINEERING, 2012, 92 (10) :891-916
[3]  
[Anonymous], 2016, ARXIV161103537
[4]  
[Anonymous], 2016, IEEE Sensor Array and Multichannel Signal Processing Workshop
[5]  
[Anonymous], 1991, 104069 NASA
[6]   Low-Dimensional Approach for Reconstruction of Airfoil Data via Compressive Sensing [J].
Bai, Zhe ;
Wimalajeewa, Thakshila ;
Berger, Zachary ;
Wang, Guannan ;
Glauser, Mark ;
Varshney, Pramod K. .
AIAA JOURNAL, 2015, 53 (04) :920-933
[7]   IEEE-SPS and connexions - An open access education collaboration [J].
Baraniuk, Richard G. ;
Burrus, C. Sidney ;
Thierstein, E. Joel .
IEEE SIGNAL PROCESSING MAGAZINE, 2007, 24 (06) :6-+
[8]   Model-Based Compressive Sensing [J].
Baraniuk, Richard G. ;
Cevher, Volkan ;
Duarte, Marco F. ;
Hegde, Chinmay .
IEEE TRANSACTIONS ON INFORMATION THEORY, 2010, 56 (04) :1982-2001
[9]   An 'empirical interpolation' method: application to efficient reduced-basis discretization of partial differential equations [J].
Barrault, M ;
Maday, Y ;
Nguyen, NC ;
Patera, AT .
COMPTES RENDUS MATHEMATIQUE, 2004, 339 (09) :667-672
[10]   A Survey of Projection-Based Model Reduction Methods for Parametric Dynamical Systems [J].
Benner, Peter ;
Gugercin, Serkan ;
Willcox, Karen .
SIAM REVIEW, 2015, 57 (04) :483-531