Search for: All records

Additive manufacturing systems are being deployed on a cloud platform to provide networked manufacturing services. This article explores the value of interconnected printing systems that share process data on the cloud in improving quality control. We employed an example of quality learning for cloud printers by understanding how printing conditions impact printing errors. Traditionally, extensive experiments are necessary to collect data and estimate the relationship between printing conditions vs. quality. This research establishes a multi-printer co-learning methodology to obtain the relationship between the printing conditions and quality using limited data from each printer. Based on multiple interconnected extrusion-based printing systems, the methodology is demonstrated by learning the printing line variations and resultant infill defects induced by extruder kinematics. The method leverages the common covariance structures among printers for the co-learning of kinematics-quality models. This article further proposes a sampling-refined hybrid metaheuristic to reduce the search space for solutions. The results showed significant improvements in quality prediction by leveraging data from data-limited printers, an advantage over traditional transfer learning that transfers knowledge from a data-rich source to a data-limited target. The research establishes algorithms to support quality control for reconfigurable additive manufacturing systems on the cloud. 

The integration of cyber-physical systems (CPS) has been extremely advantageous to society, it merges the attention of cybersecurity for vehicles as a timely concern as a matter of public and individual. The failure of any vehicle system could have a serious impact on vehicle control and cause undesired consequences. With the growing demand for security in CPS, there are few hands-on labs/modules available for training current students, future engineers, or IT professionals to understand cybersecurity in CPS. This study describes the execution of a free security testbed to replicate a vehicle’s network system and the implementation of this testbed via hands-on lab designed to introduce concepts of vehicle control systems. The hands-on lab simulates insider threat scenarios where students had to use can-utils toolkits and SavvyCAN to send, modify, and capture the network packet and exploit the system vulnerability threats such as replay attacks and fuzzing attacks on the vehicle system. We conducted a case study with 21 university-level students, and all students completed the hands-on lab, pretest, posttest, and a satisfaction survey as part of a non-graded class assignment. The experimental results show that most students were not familiar with cyber-physical systems and vehicle control systems and never had the chance to do any hands-on lab in this field before. Furthermore, students reported that the hands-on lab helped them learn about CAN-bus and rated high scores for enjoyment. We discussed the design of an affordable tool to teach about vehicle control systems and proposed directions for future work.