A comparison of physical models with Kalman filtering to deep learning models for complex system state estimation and prediction

Spencer Pollard; Samuel B. Siewert

doi:10.1117/12.3052041

A comparison of physical models with Kalman filtering to deep learning models for complex system state estimation and prediction

California State University, Chico

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Original language	English
Title of host publication	Signal Processing, Sensor/Information Fusion, and Target Recognition XXXIV
Editors	Ivan Kadar, Erik P. Blasch, Lynne L. Grewe
Publisher	SPIE
ISBN (Electronic)	9781510687479
DOIs	https://doi.org/10.1117/12.3052041
State	Published - 2025
Externally published	Yes
Event	Signal Processing, Sensor/Information Fusion, and Target Recognition XXXIV 2025 - Orlando, United States Duration: Apr 14 2025 → Apr 16 2025

Publication series

Name	Proceedings of SPIE - The International Society for Optical Engineering
Volume	13479
ISSN (Print)	0277-786X
ISSN (Electronic)	1996-756X

Conference

Conference	Signal Processing, Sensor/Information Fusion, and Target Recognition XXXIV 2025
Country/Territory	United States
City	Orlando
Period	4/14/25 → 4/16/25

ASJC Scopus Subject Areas

Electronic, Optical and Magnetic Materials
Condensed Matter Physics
Computer Science Applications
Applied Mathematics
Electrical and Electronic Engineering

Keywords

complex system
deep learning
double pendulum
Kalman filter
physics inspired machine learning
recurrent neural networks
YOLOv8

Access to Document

10.1117/12.3052041

Cite this

Pollard, S., & Siewert, S. B. (2025). A comparison of physical models with Kalman filtering to deep learning models for complex system state estimation and prediction. In I. Kadar, E. P. Blasch, & L. L. Grewe (Eds.), Signal Processing, Sensor/Information Fusion, and Target Recognition XXXIV Article 1347914 (Proceedings of SPIE - The International Society for Optical Engineering; Vol. 13479). SPIE. https://doi.org/10.1117/12.3052041

A comparison of physical models with Kalman filtering to deep learning models for complex system state estimation and prediction. / Pollard, Spencer; Siewert, Samuel B.
Signal Processing, Sensor/Information Fusion, and Target Recognition XXXIV. ed. / Ivan Kadar; Erik P. Blasch; Lynne L. Grewe. SPIE, 2025. 1347914 (Proceedings of SPIE - The International Society for Optical Engineering; Vol. 13479).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Pollard, S & Siewert, SB 2025, A comparison of physical models with Kalman filtering to deep learning models for complex system state estimation and prediction. in I Kadar, EP Blasch & LL Grewe (eds), Signal Processing, Sensor/Information Fusion, and Target Recognition XXXIV., 1347914, Proceedings of SPIE - The International Society for Optical Engineering, vol. 13479, SPIE, Signal Processing, Sensor/Information Fusion, and Target Recognition XXXIV 2025, Orlando, United States, 4/14/25. https://doi.org/10.1117/12.3052041

Pollard S, Siewert SB. A comparison of physical models with Kalman filtering to deep learning models for complex system state estimation and prediction. In Kadar I, Blasch EP, Grewe LL, editors, Signal Processing, Sensor/Information Fusion, and Target Recognition XXXIV. SPIE. 2025. 1347914. (Proceedings of SPIE - The International Society for Optical Engineering). doi: 10.1117/12.3052041

Pollard, Spencer ; Siewert, Samuel B. / A comparison of physical models with Kalman filtering to deep learning models for complex system state estimation and prediction. Signal Processing, Sensor/Information Fusion, and Target Recognition XXXIV. editor / Ivan Kadar ; Erik P. Blasch ; Lynne L. Grewe. SPIE, 2025. (Proceedings of SPIE - The International Society for Optical Engineering).

@inproceedings{2e442f6df79b44cf9363735ff3a9785b,

title = "A comparison of physical models with Kalman filtering to deep learning models for complex system state estimation and prediction",

keywords = "complex system, deep learning, double pendulum, Kalman filter, physics inspired machine learning, recurrent neural networks, YOLOv8",

author = "Spencer Pollard and Siewert, \{Samuel B.\}",

note = "Publisher Copyright: {\textcopyright} 2025 SPIE.; Signal Processing, Sensor/Information Fusion, and Target Recognition XXXIV 2025 ; Conference date: 14-04-2025 Through 16-04-2025",

year = "2025",

doi = "10.1117/12.3052041",

language = "English",

series = "Proceedings of SPIE - The International Society for Optical Engineering",

publisher = "SPIE",

editor = "Ivan Kadar and Blasch, \{Erik P.\} and Grewe, \{Lynne L.\}",

booktitle = "Signal Processing, Sensor/Information Fusion, and Target Recognition XXXIV",

}

TY - GEN

T1 - A comparison of physical models with Kalman filtering to deep learning models for complex system state estimation and prediction

AU - Pollard, Spencer

AU - Siewert, Samuel B.

PY - 2025

Y1 - 2025

KW - complex system

KW - deep learning

KW - double pendulum

KW - Kalman filter

KW - physics inspired machine learning

KW - recurrent neural networks

KW - YOLOv8

UR - https://www.scopus.com/pages/publications/105010688815

UR - https://www.scopus.com/inward/citedby.url?scp=105010688815&partnerID=8YFLogxK

U2 - 10.1117/12.3052041

DO - 10.1117/12.3052041

M3 - Conference contribution

AN - SCOPUS:105010688815

T3 - Proceedings of SPIE - The International Society for Optical Engineering

BT - Signal Processing, Sensor/Information Fusion, and Target Recognition XXXIV

A2 - Kadar, Ivan

A2 - Blasch, Erik P.

A2 - Grewe, Lynne L.

PB - SPIE

T2 - Signal Processing, Sensor/Information Fusion, and Target Recognition XXXIV 2025

Y2 - 14 April 2025 through 16 April 2025

ER -

A comparison of physical models with Kalman filtering to deep learning models for complex system state estimation and prediction

Publication series

Conference

ASJC Scopus Subject Areas

Keywords

Access to Document

Other files and links

Cite this