AUTOHIAM: Automated Robotic System for Hydrogel Infusion Additive Manufacturing

AUTOHIAM is an affordable and flexible vision-based robotic system that automates Hydrogel Infusion Additive Manufacturing.

Paper | Project | Video | Documentation

Introduction

AUTOHIAM is an open-source robotic platform that automates the entire HIAM workflow. It enables chemists to run multi-cycle, multi-sample experiments 24/7 without programming. AUTOHIAM uses vision-based pose estimation with ArUco markers and cuRobo-based GPU-accelerated motion planning to perform collision-free sample transfer. It supports liquid handling via low-cost peristaltic pumps, robot-assisted drying, and flexible obstacle-aware planning. The system interacts with chemists through natural language to configure experiments and can be monitored remotely. Tested on Gen2 metal-ceramic fabrication, AUTOHIAM achieves sample quality comparable to manual execution while improving safety, scalability, and throughput.

Citation

@misc{zhuang2025autohiam,
  author       = {Zhuang, Shengyang and Yee, Daryl},
  title        = {AUTOHIAM: Automated Robotic System for Hydrogel Infusion Additive Manufacturing (Gen2 \& Gen3)},
  year         = {2025},
  howpublished = {\url{https://epfl-alchemy.github.io/autohiam/}},
  note         = {Research project developed at ALCHEMY, École Polytechnique Fédérale de Lausanne (EPFL)},
}

Contact

Prof. Daryl Yee: daryl.yee@epfl.ch