Session: VIB-15-01: Machine Learning Applications in Vibrations and Dynamics

Paper Number: 90118

90118 - Progress Towards Data-Driven High-Rate Structural State Estimation on Edge Computing Devices 

Structures operating in high-rate dynamic environments, such as hypersonic vehicles, orbital space infrastructure, and blast mitigation systems, require microsecond (µs) decision-making. Advances in real-time sensing, edge-computing, and high-bandwidth computer memory are enabling emerging technologies such as High-rate structural health monitoring (HRSHM) to become more feasible. Due to the sub-second time restrictions such systems operate under, a target of 1 millisecond (ms) from event detection to decision-making is set as the goal to enable HRSHM. With minimizing latency in mind, a data-driven method that relies on acceleration time-series measurements taken in real-time to infer the state of the structure is investigated in this preliminary work. A methodology for deploying LSTM-based state estimators for structures using subsampled time-series vibration data is presented. The proposed state estimator is deployed to an embedded real-time device and the achieved accuracy along with system timing are discussed. The proposed approach has shown potential for high-rate state estimation as it provides sufficient accuracy for the considered structure while a time-step of 2.5 ms is achieved. The Contributions of this work are twofold: 1) a framework for deploying LSTM models in real-time for high-rate state estimation, 2) an experimental validation of LSTMs running on a real-time computing system.

Presenting Author: Joud Satme University of South Carolina

Presenting Author Biography: Joud Satme Is a graduate of electrical engineering with a background in control systems and sensor development. Satme's work revolves around high-rate machine learning in structural health monitoring applications.

Authors:

Joud Satme University of South Carolina
Daniel Coble University of South Carolina
Braden Priddy University of South Carolina
Austin Downey University of South Carolina
Jason Bakos University of South Carolina
Gurcan Comert Benedict College