Robust Affordable 3D Haptic Sensation via Learning Deformation Patterns
This addresses the challenge of making haptic sensing more affordable and robust for robotics applications, though it appears incremental by optimizing sensor placement rather than introducing a new paradigm.
The paper tackles the problem of inferring haptic forces on 3D structures from internal deformations using a small number of sensors, achieving high-precision force inference with as few as 9 sensors and 8 mm localization precision on a robotic limb.
Haptic sensation is an important modality for interacting with the real world. This paper proposes a general framework of inferring haptic forces on the surface of a 3D structure from internal deformations using a small number of physical sensors instead of employing dense sensor arrays. Using machine learning techniques, we optimize the sensor number and their placement and are able to obtain high-precision force inference for a robotic limb using as few as 9 sensors. For the optimal and sparse placement of the measurement units (strain gauges), we employ data-driven methods based on data obtained by finite element simulation. We compare data-driven approaches with model-based methods relying on geometric distance and information criteria such as Entropy and Mutual Information. We validate our approach on a modified limb of the Poppy robot and obtain 8 mm localization precision.