Anthropomorphic hand learns to crawl and manipulate objects independently
Fingers as Legs: Learning Self-Supported Locomotion and Manipulation with an Anthropomorphic Hand
Robotics
Summary
Robotic hands usually use fingers just for grabbing things, but this work shows a hand that can also walk around using its fingers like legs. The authors taught the hand through simulation how to move, steer, and recover from falls while supporting its own weight. On the real robot, the hand can crawl without any wires, press keyboard keys, and push objects to targets. This shows a small robot hand that moves itself and interacts without needing extra legs or wheels.
What this means in practice
- •For robotic system developers: Develop compact robots that use finger-like limbs for both moving and handling objects without separate movement parts.
- •For assistive technology designers: Create single-device robotic helpers capable of navigating and interacting with environments using one hand-shaped manipulator.
Authors
Amirhossein Kazemipour, Hehui Zheng, Robert Katzschmann
Abstract
A walking robotic hand must use the same fingers to move its body, support its weight, and interact with the environment. We show how an anthropomorphic hand can learn these skills while retaining its finger design and position controller. Onboard power and computation make the platform self-contained. Our reinforcement learning approach accounts for the hand's unequal fingers, with training in a simulator calibrated from hardware measurements. In simulation, the hand moves faster with our reward formulation than with tuned rewards originally designed for quadrupeds. On hardware, task-specific policies enable untethered crawling, steering, and fall recovery. While supporting its own weight, the hand also executes successive keyboard commands without vision and pushes an object to targets using overhead visual feedback. These results demonstrate a compact mobile manipulator that reuses its fingers for locomotion and interaction, without a separate locomotion mechanism.