Online detection of cyber‐incidents in additive manufacturing systems via analyzing multimedia signals

Abstract

Additive manufacturing (AM) or 3D printing is an emerging manufacturing technology that plays a growing role in both industrial and consumer settings. However, security concerns of AM systems have been raised among researchers. In this paper, we present an online detection mechanism for the malicious attempts on AM systems, which taps into both audio and video signals collected during the printing process. For audio signals, we propose to monitor the shift of patterns in the spectrogram and dominant frequencies via a control chart designed based on the Wasserstein metric. For video signals, we propose to monitor the change in the reconstructed path of the extruder via a Hausdorff metric. We then show the effectiveness of our methods in a case study using an Ender 3D printer, where the cyber‐incidence of altering the internal fill density can be easily identified in an online manner.

A tensor‐based hierarchical process monitoring approach for anomaly detection in additive manufacturing

“…Let  ∈ ℝ 𝐼 1 ×𝐼 2 ×𝐼 3 ×𝐼 4 ×𝑁 denote the tensor of 𝑁 in-control samples  𝑘 ; 𝑘 = 1, 2, … , 𝑁. Using proposition 1, we can argue that the basis matrices Â, B, Ĉ, and D obtained through Tucker decomposition of  provide the same solutions to Equation ( 8) 5 and columnwise orthogonality for 𝑨, 𝑩, 𝑪, 𝑫, 𝑬 where 𝑬 is the basis matrix corresponding to the last dimension (sample) of tensor ,  is the extracted core tensor and 𝑡 𝑝𝑞𝑟𝑙ℎ is the (𝑝, 𝑞, 𝑟, 𝑙, ℎ)th element of core tensor  . Using proposition 1, given  and Ê the core tensor (𝑘) in Equation ( 8) can be obtained by…”

“…A major field of research regards the use of in‐line data coming from in situ sensors to automatically detect process anomalies and the onsets of defects in the part. Methods in this field are commonly referred to as “in situ process monitoring.” 3–6 A key driver for in situ monitoring in AM is the ability to anticipate unstable and out‐of‐control states during the process rather than relying on costly and time‐consuming post‐process nondestructive tests. Combining this ability with reactive, adaptive, or corrective actions could help one mitigate undesired part‐to‐part variations and keep defect rates compliant with stringent quality requirements.…”

Cybersecurity Issues in Motion Control – An Overview of Challenges

“…As a rule, in these investigations exteroceptive sensors were used to compare the commanded with the executed trajectory. For example, two approaches based on the analysis of multimedia signals for the detection of cyber-attacks in additive manufacturing are proposed in [25]. One method recognizes cyber-incident using spectral analysis of audio signal and the other carries out the comparison of the path reconstructed from video with the commanded path.…”