The goal of this project is to create a robotic canopy that can extend, contract, bend, twist in a dynamic and socially interactive way because of the inherent flexibility of origami patterns. This robotic origami canopy is the first step towards the empirical investigation of Socially Interactive Robotic Environments (SIREs), which refers to built environments embedded with socially expressive architectural robotics that make the physical space reconfigurable and socially interactive as if our friends, companions, partners, etc. This idea of “a room or space that comes to life and becomes a social agent (e.g., friends, caregivers, and even lovers)” has been widely discussed in sci-fi literature, but rarely explored empirically as a research topic in academia. To explore how people perceive and interact with SIREs, we need to first make a SIRE. This is the initial inspiration for this robotic origami canopy project: designing, engineering, and constructing a large-scale space-making robotic canopy.
Funded by: Arts at Tech Catalyst Grant, “Enter, Touch, and Feel”: Design and Development of a Socially Interactive, Continuum Origami Robotic Structure in Room Scale
PI: Yixiao Wang
Co-PI: Abigale Stangl, Daniel Baerlecken
Student Research Assistants: Olaf Kamperman, Erin Grissom, Hudson Musnicki, Yifan Hou, Junzhe Yan, Qi Ting Yu
Publications: Olaf Kamperman, Erin Grissom, and Hudson Musnicki. 2025. Designing a Socially Interactive Robotic Origami Canopy for Augmenting Art Viewing Experiences. In Proceedings of the 2025 ACM/IEEE International Conference on Human-Robot Interaction (HRI ’25). IEEE Press, 1926–1929.
Video Publication: https://dl.acm.org/doi/10.5555/3721488.3721815
This project serves as the pilot project supporting PI’s NSF CAREER grant application.
Poster Presentation at the ACM HRI 2025 Conference in Melbourne, Australia:



Exhibition at the Robert C. Williams Museum of Papermaking – Georgia Tech (Feb. 24 to Feb 28 during the Workshop on Designing for Tactile and Embodied Learning)






Here is the journey of our design research exploration:
Origami Pattern Explorations: What kind of origami patterns can realize the dynamic, playful movement of a robotic canopy? We explored a variety of different origami patterns that can potentially enable bend, twist, extend, contract, and many other shape-changing movements.





Material Exploration I: For constructing a relatively small-scale origami canopy (covering one person), what materials can work?










Scale Explorations: How big this robotic origami canopy should be?


Material Explorations II: For constructing a relatively large-scale origami canopy (covering at least 3 people), what materials can work? Following the trajectory of our previous material exploration, we first tried different kinds of fabric since an origami pattern usually applies to one piece of material. We used tapes, glues, and other materials to address the folds of the original origami pattern.




Structure Explorations: Based on the material and scale explorations above, we realized that making the whole canopy out of one piece of material may probably compromise its structural stability. Thus, reconfigurable and weight-bearing structures should be created. We first experimented with plastic cloth hangers because of their lightweight, cheap price, and triangular shape (which fits well with our origami patterns). Then we moved on to the high-fidelity materials which are carbon fiber tubes covered by Mylar sheets.
Plastic Cloth Hanger Explorations:






Carbon Fiber Tube Explorations






Actuation System Explorations: As a shape-changing structure, how do we actuate the spatial reconfigurations? Based on the structural, material, scale, and origami explorations above, we designed two kinds of actuation systems: the “tendon-driven system” and the “scissor-driven system.”
The Tendon-driven System









The Scissor-driven System
Compared with the tendon-driven system, the scissor system have the following advantages:
1) This system is weight-bearing. It can support the weight of relatively heavy canopy structure, which is inevitable as the scale of the canopy grows larger;
2) This system can be fully embedded in the robotic canopy itself, and thus, it is portable and does not require on-site installations. The tendon driven system, on the other hand, requires on-site installations wherever the exhibition is going to be.
The novelty of our scissor-driven system is that, it has infinite degrees of freedom on the XY plane (which means it can bend freely from left to right). In other words, this is a continuum robot mechanism. It tool quite a lot of experimentation to come up with a mechanism that is:
1) vertically weight-bearing (to support the weight of large-scale robotic canopy);
2) horizontally bendable with infinite degrees of freedom (to enable playful and socially expressive robotic canopy movement);
3) horizontally extendable and contractable (to support the folding and unfolding of the origami pattern).