Friction Variability in Planar Pushing Data: Anisotropic Friction and Data-Collection Bias

Friction plays a key role in manipulating objects. Most of what we do with our hands, and the most of what robots do with their grippers, is based on the ability to control frictional forces. This letter aims to better understand the variability and predictability of planar friction. In particular, we focus on the analysis of a recent dataset on planar pushing by [K.-T. Yu, M. Bauza, N. Fazeli, and A. Rodriguez, More than a Million Ways to Be Pushed: A High-Fidelity Experimental Data Set of Planar Pushing, in Proceeding of the IEEE/RSJ Internatinonal Conference on Intelligent Robots and Systems, 2016, pp. 30-37.] devised to create a data-driven footprint of planar friction. We show in this letter how we can explain a significant fraction of the observed unconventional phenomena, e.g., stochasticity and multimodality, by combining the effects of material nonhomogeneity, anisotropy of friction and biases due to data collection dynamics, hinting that the variability is explainable but inevitable in practice. We introduce an anisotropic friction model and conduct simulation experiments comparing with more standard isotropic friction models. The anisotropic friction between the object and supporting surface results in convergence of initial condition during the automated data collection. Numerical results confirm that the anisotropic friction model explains the bias in the dataset and the apparent stochasticity in the outcome of a push. The fact that the data collection process itself can originate biases in the collected datasets, resulting in deterioration of trained models, calls attention to the data collection dynamics.