Researchers at ETH Zurich used reinforcement learning to teach the WUJI robot hand to move on its fingers, in a paper titled "Fingers as Legs" by Kazemipour, Zheng and Katzschmann.
The hand can crawl, steer, recover from a fall, press keyboard keys and push objects.
Source: roboticsandautomationnews.com, reported 10-16. https://roboticsandautomationnews.com/2026/10/16/eth-zurich-scientists-turn-wuji-robotic-hand-into-self-propelled-mobile-manipulator/105991/