Extrinsic Dexterity through Active Slip Control using Deep Predictive Models


We present a machine learning methodology for actively controlling slip, in order to increase robot dexterity. Leveraging recent insights in Deep Learning, we propose a Deep Predictive Model that uses tactile sensor information to reason about slip and its future influence on the manipulated object. This information can then be used to precisely manipulate various objects within the robots hand using external perturbations imposed by gravity or acceleration. We show in a set of experiments that this approach can be used to increase a robots repertoire of skills.

Accepted for International Conference on Robotics and Automation, IEEE.