Abstract:
Dynamic touch sensing has shown potential for multiple tasks. In this talk, I will present how we utilize dynamic touch sensing to perceive particles inside a container with two tasks: classification of the particles inside a container and property estimation of the particles inside a container.

First, we try to recognize what is inside the container only by touch sensing. Humans solved this task by shaking the container and feeling the vibrations of the particles. To sense the fingertip vibration signals, we design a high-speed GelSight with both high spatial resolution and high temporal resolution. By shaking the container and extracting the spectrum features, we are able to classify five different particles. Multiple comparison experiments show that both high spatial resolution and high temporal resolution of tactile signals help with this task.

Then we step forward to estimate fundamental particle properties instead of classification, specifically, content mass, content volume, particle size, and particle shape. We design a sequence of robot actions to interact with the container. Based on physical understanding, we extract static force/torque value from the F/T sensor, vibration-related features, and topple-related features from the high-speed GelSight tactile sensor to estimate those four particle properties. Experiments show that our method works well for both seen and unseen particles.

Committee:
Wenzhen Yuan
Chris Atkeson
Oliver Kroemer
Kevin Zhang