ENBIS Spring Meeting 2018

4 – 6 June 2018; Florence, Italy Abstract submission: 17 November 2017 – 20 April 2018

AKM2D : An Adaptive Framework for Online Sensing and Anomaly Detection

5 June 2018, 16:30 – 16:55


Submitted by
Kamran Paynabar
Kamran Paynabar (School of Industrial and Systems Engineering), Hao Yan (Sch Compt Infor & Dec Sys Engr, Arizona State University), Jianjun Shi (School of Industrial and Systems Engineering)
In point-based sensing systems such as coordinate measuring machines (CMM) and laser ultrasonics where complete sensing is impractical due to the high sensing time and cost, adaptive sensing through a systematic exploration is vital for online inspection and anomaly detection. Most of existing sequential sampling methodologies focus on reducing the overall fitting error for the entire sampling space. However, in many anomaly detection applications, the main goal is to accurately detect and estimate sparse anomalous regions. In this paper, we develop a novel framework named Adaptive Kernelized Maximum-Minimum Distance (AKM2D) to speed up the inspection and anomaly detection process through an intelligent sequential sampling scheme integrated with fast estimation and detection. The proposed method balances the sampling efforts between the space filling sampling (exploration) and focused sampling near the anomalous region (exploitation). The proposed methodology is validated by conducting simulations and a case study of anomaly detection in composite sheets using a guided wave test.

Return to programme