Most sensing systems add a discrete electronic layer to an otherwise passive structure. We ask a different question: what can be measured when the material itself is the instrument?
Our work combines soft conductive composites, geometric mechanics, and sparse inference. Rather than optimizing a single sensor, we design the distribution of material response across an object and recover useful state from that response.
Current questions
- How much spatial information can a small number of electrical measurements retain?
- Can reversible material changes record short environmental histories?
- Which fabrication constraints improve, rather than limit, inference?
The aim is not to make every object “smart.” It is to build systems in which computation begins with an honest account of matter.