A Universal Edge-Cloud Architecture for Autonomous 3D Printing
Bing, Zhang; Lomas, Johnny Austin; Puthanveetil Satheesan, Sandeep; Berry, Matthew John; Diao, Ying
- Publisher
- Zenodo
- Language
- en
Abstract
As 3D printing moves toward full autonomy, the need for a universal, scalable, and intelligent architecture becomes increasingly critical. Modern additive manufacturing faces persistent barriers: fragmented device interfaces, lack of remote operability, ad hoc parameter tuning, and insufficient integration between physical processes and data-driven control. As additive manufacturing evolves toward more complex applications such as structural color fabrication, there is a growing demand for a universal, intelligent control framework capable of integrating diverse printer hardware, automating parameter optimization, and managing experimental data at scale. Our work presents a novel universal edge-cloud architecture that combines edge-side hardware adaptors with a cloud-based backend to support real-time control, intelligent optimization, and robust data management in the domain of autonomous 3D print. A centralized, user-friendly web interface allows users to configure and launch print jobs (campaigns), define experimental parameters, and review historical results. The cloud backend is built with scalable microservices, including a high-performance RabbitMQ-based messaging system for edge-cloud communication, MongoDB for structured storage of experimental metadata, print records and Clowder for managing archival files and analysis reports. Distributed image analysis and machine learning-based prediction services run in parallel to optimize print parameters. This supports autonomous closed-loop experimentation across large-scale campaigns. To ensure flexibility and hardware independence, print jobs are abstracted using a custom PCP (Parameterized Control Protocol) file format. PCP files encapsulate device-specific instructions along with structured metadata describing print sequences, geometric layout, and adjustable parameters. This abstraction enables batch-based execution and seamless orchestration across printers running different configurations. The system exchanges messages with 3D printers via a custom-developed edge adaptor deployed to the near printers. We validate the system through an application in structural color 3D printing, focusing on the relationship between color properties—represented in the HSV color space—and key printing parameters such as pressure, speed, and bed temperature. The system was deployed on the UIUC Radiant cloud cluster and tested across multiple print shapes defined by PCP files on the 3D printers using Marlin firmware. To efficiently explore the parameter space and enhance predictive performance, we incorporate a Bayesian optimizer into the machine learning workflow. The system’s adaptability and modular design enable scalable, reproducible experimentation across a wide range of hardware, demonstrating its potential as a universal platform for autonomous 3D printing.
Full text
Bing Zhang¹, Johnny Austin Lomas², Sandeep Puthanveetil Satheesan¹, Matthew Berry¹, Ying Diao² ¹National Center for Supercomputing Applications; ²Department of Chemical and Biomolecular Engineering, College of Liberal Arts and Sciences University of Illinois Urbana-Champaign A Universal Edge-Cloud Architecture for Autonomous 3D Printing ACKNOWLEDGEMENTS ● This material is based upon work supported by the Illinois Computes program. ● The experiments presented in this work were obtained using the NCSA Radiant cloud resources. ● This research is in part sponsored by NSF under award number 2119172. OUR SOLUTION ● We propose a hybrid edge-cloud framework that unifies heterogeneous components and addresses autonomy challenges. ○ edge computing role: Edge nodes provide low-latency feedback loops, local monitoring, and real-time decision making at the machine level. ○cloud computing role: Cloud platforms enable scalable computation, data storage, and system-wide coordination. ● This system architecture enables scalable, reliable, and adaptable autonomous 3D printing networks across domains. ○ The system was deployed on the UIUC Radiant cloud cluster. ○ The system supports UIUC researchers in conducting structural color 3D printing experiments. ● Integrated with printers running Marlin firmware. ○ Marlin firmware is one of the most popular open-source software interpreting G-code and controlling the printer to create 3D objects. ○ Marlin Firmware github: Fork 19.5k, Star 17k. INTRODUCTION ●Growing Importance of 3D Printing: A cornerstone of the future digital economy, transforming industries such as healthcare, construction, and aerospace, and beyond. Its integration with AI, robotics, and smart materials drives the shift toward intelligent, adaptive, and decentralized manufacturing. ●Autonomy 3D Printing Challenges: ○closed-loop control: Without human intervention, system monitors continuous feedback from sensors to adapt/adjust parameters, correct errors, and ensure print quality. ○coordination: Multi-device and multi-task environments demand synchronized decision making across distributed systems. ○resource demand: Machine learning, simulation, and optimization tasks exceed the computational limits of local computing. ●Heterogeneous Environment: Diverse printers, sensors, controllers, and software platforms and user print tasks complicate integration and interoperability. ●A Universal Platform to accommodate a variety of user tasks. CAMPAIGN EXPERIMENTS MANAGEMENT User Management New Campaign Management ● Option to set initial parameters. ● Display the content of PCP file. ● Dynamic Grid Sizing: Loads PCP file, calculates bounding box and movements, and visualizes the print bed with dynamically sized grid cells. ● Colors grid cells in real time as each cell is printed. Previous Campaign Experiment Management Print Image Archival Management ● Print images of each campaign are automatically collected and sent to the archival service. ● Each campaign becomes a dataset containing all its images. Users can click a colored cell to access its print image directly. 3D PRINTER WE USE Lulzbot taz pro 3D printer in Prof. Ying Diao’s lab. CONCLUSIONS & FUTURE WORK System validation via structural color 3D printing application, analyzing how HSV color properties relate to printing parameters like pressure, speed, and bed temperature. The system was deployed on the UIUC Radiant cloud cluster and tested on multiple PCP-defined prints using Marlin firmware-based 3D printers. We conducted closed-loop campaigns without human intervention, guided by a Bayesian optimizer to efficiently explore the parameter space. Its modular, adaptable design enables scalable, reproducible experiments, demonstrating its potential as a universal platform for autonomous 3D printing. Future work includes optimizing system performance and reliability, and evaluating diverse 3D printing tasks. HIGH-LEVEL SYSTEM OVERVIEW Adapter ● Running on a machine near to 3D printer. ○ Receives command messages and translates them into G-code compatible with the Marlin firmware. ○ Quick check of print image quality. ○ Automatically cleans the nozzle after each cell print. ○ Store image files to archival service and further analysis. A Centralized, User-friendly Web Application ● Provides an administrative interface for managing remote 3D printers. ○ User account management. ○ Browse historical campaign print results. ○ Start new campaigns with customized print parameters. Cloud-Based Scalable Microservices ● User-built docker images are managed by an orchestration toolkit, which scales multiple containers across a cloud-based cluster of machines. ● High-performance RabbitMQ-based messaging system for edge-cloud communication ● MongoDB for structured storage of experimental metadata, print records. ● Archival service (Clowder) for managing archival files and analysis reports. FAILURE TOLERANCE ● Replicated microservices for reliability. ● Persistent task commands in messaging system. ● Alerts on 3D printer disconnections. ● Database maintains all print records.
