What might our surroundings look like, in a world of ubiquitous robots? This dissertation reimagines the distinction between robots and spaces, introducing an architectural integration paradigm where robotic systems become one with our built environment. Rather than existing as distinct entities, these shape-changing surfaces can be integrated into walls and floors, enabling environmental adaptations in response to contextual requirements. While prior work has demonstrated the promise of such dynamic systems, pre-existing strategies for fabrication and actuation often fall short when it comes to constructing shape-changing interfaces that span larger areas.

This research addresses three fundamental challenges in realizing room-scale robotic surfaces: the development of scalable materials and fabrication methodologies, the implementation of natural human-surface interaction paradigms, and the engineering of large-scale shape-transformation mechanisms. Specifically, we introduce: (1) a technique for creating "flat-fabricated" electro-mechanical components via a PCB fabrication pipeline; (2) a geometric state estimation system that enables surfaces to sense and reconstruct their own shape during human interaction; (3) a constraint-driven actuation approach that achieves robust transformations with minimal power requirements; and (4) a pair of design and elicitation studies that uncover natural interaction techniques for human-robot collaboration in adaptive environments. We conclude with a discussion of how these contributions can lead to fluent collaboration between people and their surroundings.

187 pages

Thesis Committee:
Scott E. Hudson (Co-Chair)
Alexandra Ion (Co-Chair)
Nikolas Matelaro
Nadya Peek (University of Washington)

Brad A. Myers, Head, Human-Computer Interaction Institute
Martial Hebert, Dean, School of Computer Science

Return to: SCS Technical Report Collection
School of Computer Science homepage

This page maintained by reports@cs.cmu.edu