Learning Latent Space Dynamics for Tactile Servoing
share
In order to achieve a dexterous robotic manipulation, we need to equip our robot with a tactile feedback capability, i.e. the ability to drive action based on tactile sensing. In this paper we specifically address the challenge of tactile servoing, i.e. given the current tactile sensing and a target/goal tactile sensing --for example being memorized from a successful task execution in the past--, what is the action that will bring the current tactile sensing to move closer towards the target tactile sensing at the next time step. We develop a data-driven approach to acquire a dynamics model for tactile servoing by learning from demonstration. Moreover, our method represents the tactile sensing information as to lie on a surface --or a 2D manifold-- and perform a manifold learning, making it applicable to any tactile skin geometry. As a proof of concept, we evaluate our method on a robot equipped with a tactile finger. A video demonstrating our approach can be seen in https://youtu.be/5EJSAoUO0E0
READ FULL TEXT