A comprehensive and high-fidelity dataset of planar pushing experiments. The dataset contains timestamped poses of a pusher and a pushed object, as well as forces at the interaction.

In this dataset we recorded robot motion, object motion and forces and torques at all the contacts for some prehensile pushing experiments. Now, the question we ask is, what is a good contact dynamics model to explain this data and to simulate in-hand manipulations?

In this work we captured a dataset of slow-motion actions in a convenience store setting. We analyzed this video through the lenses of different manipulation taxonomies.

The dataset used to evaluate our work on realtime state estimation with tactile and visual sensing in planar pushing.

An accurate, diverse, real-world dataset of planar pushing that comprises an extensive set of pushed objects and RGBD images of the interactions.

The dataset used to evaluate our work on realtime state estimation in inserting a suction-held object.