Non-linear vibration signal compensation technique for UAV-deployable sensor packages with edge computing

Joud N. Satme; Daniel Coble; Hung-Tien Huang; Austin R. J. Downey Jr.; Jason D. Bakos

doi:10.1117/12.2658563

18 April 2023 Non-linear vibration signal compensation technique for UAV-deployable sensor packages with edge computing

Joud N. Satme, Daniel Coble, Hung-Tien Huang, Austin R. J. Downey Jr., Jason D. Bakos

Proceedings Volume 12486, Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems 2023; 124861M (2023) https://doi.org/10.1117/12.2658563
Event: SPIE Smart Structures + Nondestructive Evaluation, 2023, Long Beach, California, United States

Abstract

For rapid assessment of infrastructure, the use of minimally invasive sensors that can be deployed remotely using autonomous vehicles is gaining popularity. Such systems are favorable for their ease of deployment and cost-effectiveness. Utilizing electropermanent magnets or adhesives to mount the sensors temporarily forms a barrier between the sensor and the structure being examined. This barrier creates undesirable nonlinearities and transmissibility losses that introduce errors into structural damage detection algorithms. Post-processing of signals using continuous modeling techniques from classical control theory can be applied to the collected signals to remove this error. However, post-processing creates additional analysis steps that require the signal to be taken off device. Processing the data at-the-edge prior to saving it to memory or transmitting it to a base station enables rapid assessment of infrastructure. With minimal time from signal detection to prognostics, such systems can be used in damage forecasting and infrastructure failure prevention. This preliminary work aims to develop a non-linear machine-based compensation technique that is resource and power efficient enough to be processed on-device. The proposed long short-term memory (LSTM) error-compensating network demonstrated potential by increasing the SNRdB by 9.3% and improving RMSE by approximately 20% while widening the usable lower limit of the sensor’s bandwidth from 2.78 to 1.34 Hz. The progress described in this report focuses on setting the framework for the proposed method and paves the way for a full-scale hardware implementation in the near future

Conference Presentation

Citation Download Citation

Joud N. Satme, Daniel Coble, Hung-Tien Huang, Austin R. J. Downey Jr., and Jason D. Bakos "Non-linear vibration signal compensation technique for UAV-deployable sensor packages with edge computing", Proc. SPIE 12486, Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems 2023, 124861M (18 April 2023); https://doi.org/10.1117/12.2658563

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available