Spatiotemporal G-Code Modeling for Secure FDM-based 3D Printing

Paper Details

• Title: Spatiotemporal G-Code Modeling for Secure FDM-based 3D Printing
  
• Authors: Muhammad Haris Rais, Ye Li,  Irfan Ahmed
• Publication/Conference: 12th ACM/IEEE International Conference on Cyber-Physical Systems (ICCPS)
• Publication/Presentation Date:  May 2021

Abstract

Detecting attacks on 3D printers, which construct objects by building and stacking layers according to Computer-Aided Design information, relies on creating a master profile for each unique object, then printing it in a protected environment. This is unsustainable in mass production. 

Researchers developed the Sophos system to detect attacks on the first print by using spatiotemporal G-code modeling. The system does not require prior learning for each unique object.

Sophos is scalable and supports modular upgrades to suit different printing requirements. Its design allows the detection threshold to be reduced to as low as the 3D printer’s resolution, shifting to a study of attack patterns instead of attack magnitudes.