touchdexterity.github.io - Rotating without Seeing

home > domains > touchdexterity.github.io

Description: Rotating without Seeing: Towards In-hand Dexterity through Touch

dexterous manipulation (5)

Example domain paragraphs

We present a new dexterous manipulation system design and learning pipeline for in-hand rotation using only touch. Our system is able to rotate various real-world objects around different axes without seeing them. Video speed is 1x.

Tactile information plays a critical role in human dexterity. It reveals useful contact information that may not be inferred directly from vision. In fact, humans can even perform in-hand dexterous manipulation without using vision. Can we enable the same ability for the multi-finger robot hand? In this paper, we present Touch Dexterity , a new system that can perform in-hand object rotation using only touching without seeing the object. Instead of relying on precise tactile sensing in a small region, we in

We present a new dexterous manipulation system design and learning pipeline for in-hand rotation using only touch. Our system is able to rotate various real-world objects around different axes without seeing them. The hardware is consisted of an Allegro robotic hand with 16 low-cost force-sensing resistor (FSR) sensors. We train a control policy to rotate multiple objects in the simulation through RL. Then the learned policy can directly transfer to the real and generalize to unseen, novel objects.

Links to touchdexterity.github.io (4)

• xiaolonw.github.io Xiaolong Wang
• yzqin.github.io Yuzhe Qin
• binghao-huang.github.io BinghaoHuang's Homepage
• zhaohengyin.github.io Zhao-Heng Yin