Time-Division Multimodal Tactile Perception for Physical AI and Robotic Hands
Dohyung Kim, Kyun Kyu Kim, Junhyuk Bang, Seung Hwan Ko
AI summary
Problem
Existing multimodal tactile sensors rely on bulky stacked or laterally patterned layers, causing spatial offsets, increased thickness, poor conformability on curved fingers, and slow response times that hinder real-time robotic manipulation.
Approach
The authors developed a biomimetic time-division tactile platform using a single-layer Ag-Cuâ‚‚O core-sheath nanowire network that switches between low-resistance (mechanical) and high-resistance (thermal) states via memristive transitions at 16 Hz.
Key results
- Sub-microsecond mechanical and millisecond thermal response times
- 95% object classification accuracy using time-division multimodal input
- 83% accuracy recognizing 20 household objects via wireless module
- ~1 mm spatial resolution in a scalable multi-array configuration
Why it matters
This single-layer, rapidly switching tactile skin enables seamless integration onto curved robotic grippers, advancing real-time material-aware manipulation for physical AI.
Abstract
Equipping robotic end-effectors with human-like tactile perception is crucial for dexterous manipulation, requiring simultaneous thermal and mechanical sensing at the contact interface. Conventional multimodal sensors often rely on stacked or patterned layers, which increase device thickness, reduce conformability on curved robotic fingers, and introduce response delays. To address this, we present a time-division tactile perception platform tailored for robotic applications that utilizes memristive Ag-Cu2O core-sheath nanowire networks. This ultrathin artificial skin alternates between thermal and mechanical modalities at 16 Hz via memristive transitions, mirroring the processing of biological mechanoreceptors. In the SET state, sparse silver filaments form a mechanically sensitive network. During RESET, the semiconducting Cu2O sheath provides high thermal sensitivity. Lacking reactive components, the sensor achieves sub-microsecond mechanical and millisecond thermal responses, ideal for real-time robotic feedback. A deep learning pipeline processing these time- division signals improved object classification accuracy to 95%. Using a wireless module, 20 household objects were recognized with 83% accuracy. This single-layer architecture enables direct, seamless integration onto robotic hands, laying the groundwork for multimodal tactile intelligence in physical AI.