More than a million ways to be pushed. A high-fidelity experimental dataset of planar pushing

Abstract

Pushing is a motion primitive useful to handle objects that are too large, too heavy, or too cluttered to be grasped. It is at the core of much of robotic manipulation, in particular when physical interaction is involved. It seems reasonable then to wish for robots to understand how pushed objects move.In reality, however, robots often rely on approximations which yield models that are computable, but also restricted and inaccurate. Just how close are those models? How reasonable are the assumptions they are based on? To help answer these questions, and to get a better experimental understanding of pushing, we present a comprehensive and high-fidelity dataset of planar pushing experiments. The dataset contains timestamped poses of a circular pusher and a pushed object, as well as forces at the interaction. We vary the push interaction in 6 dimensions: surface material, shape of the pushed object, contact position, pushing direction, pushing speed, and pushing acceleration. An industrial robot automates the data capturing along precisely controlled position-velocity-acceleration trajectories of the pusher, which give dense samples of positions and forces of uniform quality.We finish the paper by characterizing the variability of friction, and evaluating the most common assumptions and simplifications made by models of frictional pushing in robotics.

Omnipush: accurate, diverse, real-world dataset of pushing dynamics with RGB-D video

“…Since we presented our earlier dataset on planar pushing [1], it has been directly used for: 1) Stochastic modeling: [23,10,17] 2) Modeling from rendered images: [16] 3) Model identification: [24] 4) Learning models for control: [25,26] 5) Filtering: [27] 6) Meta-learning: [28] With this new dataset we hope to further facilitate research in learning models and control.…”

Benchmarking Simulated Robotic Manipulation Through a Real World Dataset

“…Datasets that have been collected through the use of motion capture include human pose estimation [20], interactive manipulation performed by a human [21] and sequences of labelled RGB-D images for SLAM [1]. More specifically for tasks by robot manipulators, there are motion capture datasets of robots performing planar pushing and datasets of rigid body contacts, but here the emphasis is on tracking the effects of the robot in the context of a task, not the robot itself [22], [23]. Tracking of a dexterous manipulator to overcome the reality gap is similar to what we propose, however this information isn't available as a re-usable benchmark dataset [24].…”

Exact Bounds on the Contact Driven Motion of a Sliding Object, With Applications to Robotic Pulling

“…the Peshkin bound [13]. We test the bounds over the objects in the MIT Pushing Dataset [17] and randomly generated bipods, tripods, and quadrapods. The generated n-pods were chosen to have circumcircle diameters similar to the MIT objects, roughly 0.16m.…”

“…First, pressure distributions of objects are statically indeterminant (barring the case of three-point support with known center of mass). Second, surface imperfections lead to spatial variability in both the pressure distribution and coefficient of friction [17]. Though several force-motion models for pushing exist [19,7,4], the above sources of indeterminacy ultimately lead to errors in the predicted velocity of the pushed object.…”