Blind Manipulation of Deformable Linear Objects Based on Force Information from Environmental Contacts

Other conference contribution, 2022

Humans have the ability to manipulate Deformable Linear Objects (DLOs) such as cables, with little or no visual information by relying on force torque (FT) sensing. In this work, we propose a simplified graph representation of a DLO which enables such blind manipulation by keeping the object under tension. We show that our representation can be estimated online by a Kalman filter using FT data, which can then be used to define a set of sliding and clipping manipulation primitives. Both the estimation and manipulation primitives were tested, demonstrating that force-based perception can be sufficient for robotic tasks such as cable harness production.