Tactile Sensing for Dexterous Robot Hands: A Survey
In 30 seconds
Survey of sensor modalities, coverage and learning methods for in-hand manipulation.
Research question
Which tactile modalities matter for learned manipulation?
Problem
No consensus on sensor design; datasets are incompatible.
Previous approach
Vision-only manipulation.
New approach
Taxonomy plus benchmark recommendations.
Results
Identifies coverage density and durability as the open problems.
Limitations
Survey; no new experiments.
Industry impact
Component suppliers in tactile skins are a likely M&A target category.