DextAIRity: Deformable Manipulation Can be a Breeze

• DOI: 10.15607/rss.2022.xviii.017
• Publication Date: 2022-07-02T22:49:15Z
• Authors: Zhenjia Xu; Cheng Chi; Benjamin Burchfiel; Eric Cousineau; Siyuan Feng; Shuran Song
• Keywords: FOS: Computer and information sciences; Computer Science - Robotics; 0209 industrial biotechnology; 02 engineering and technology; Robotics (cs.RO)

This paper introduces DextAIRity, an approach to manipulate deformable objects using active airflow.In contrast conventional contact-based quasi-static manipulations, Dex-tAIRity allows the system apply dense forces on out-ofcontact surfaces, expands system's reach range, and provides safe high-speed interactions.These properties are particularly advantageous when manipulating under-actuated with large surface areas or volumes.We demonstrate effectiveness of DextAIRity through two challenging object manipulation tasks: cloth unfolding bag opening.We present a self-supervised learning framework that learns effectively perform target task sequence grasping air-based blowing actions.By closed-loop formulation for blowing, continuously adjusts its direction based visual feedback in way is robust highly stochastic dynamics.We deploy our algorithm real-world three-arm evidence suggesting can improve efficiency tasks, such as unfolding, enable new applications impractical solve quasistatic manipulations (e.g., opening).Blow 1 Blow 2 3 Zoomed-in Particle-view Input observation Gripper 10°1 9 particles per layer Cloth Air flow 5m/s