Over the past decades, roboticists have developed increasingly advanced systems that can emulate some human capabilities and effectively tackle various real-world tasks. To reliably grasp, manipulate and utilize objects in their surroundings, robots should be able to detect and process tactile information, replicating the processes underpinning the human sense of touch.
Researchers at Peking University, Beijing Institute for General Artificial Intelligence and Queen Mary University of London recently developed F-TAC hand, a bio-mimetic robotic hand designed to imitate human tactile perception. This hand, introduced in a paper published in Nature Machine Intelligence, was found to outperform various existing robotic hands on tactile manipulation tasks, by collecting detailed tactile information that can be used to plan the hand’s movements and trajectories.
“This research was driven by the goal of enabling machines to manipulate objects with precision and adaptability akin to human capabilities,” Yixin Zhu, the corresponding author of the paper, told Tech Xplore. “Our inspiration to address this problem stemmed from the intricate tactile sensing capabilities observed in the human hand and the specialized neural processing within the primary somatosensory cortex, which efficiently interprets and integrates vast amounts of sensory input in real-time.”
The main goal of this recent study by Zhu and his colleagues was to develop a new robotic hand that can seamlessly integrate tactile feedback collected across its surface with sensory-motor mechanisms. As most existing robotic hands fail to do so, such a hand could perform better on various real-world manipulation tasks.
“The F-TAC Hand replicates the biological structure of the human hand by integrating 17 high-resolution tactile sensors across 70% of the palm’s surface, achieving an impressive spatial resolution of 0.1 millimeters—equivalent to 10,000 tactile pixels per square centimeter,” explained Zhu.
“A probabilistic algorithm enables the generation of diverse, human-like grasping strategies spanning 19 common grasp types. This allows the F-TAC Hand to adapt dynamically, switching strategies within just 100 milliseconds when faced with unexpected scenarios.”
The most notable advantage of the bio-mimetic hand developed by this research team is that it can pick-up tactile data across its entire surface, utilizing this data to plan its own movements. To assess its performance, Zhu and his colleagues carried out extensive experiments, during which the F-TAC Hand completed a wide range of manual tasks in real-world scenarios.
“In 600 real-world experiments, the F-TAC Hand demonstrated a significant improvement in multi-object grasping success rates, increasing from 53.5% to 100% compared to systems lacking tactile-informed feedback,” said Zhu. “Potential practical applications include surgical assistance, high-precision assembly, aerospace operations, and emergency response scenarios that demand exceptional operational precision.”
The researchers’ initial findings highlight the advantages of the new hand they developed, suggesting that it could help to boost the sensory and motor capabilities of robots. In the future, the hand could be improved further, integrated with different humanoid robots, and tested in further experiments to assess its potential for specific applications.
“We now plan to continue deepening the integration of tactile sensing and robotic control, exploring more intelligent somatosensory interaction paradigms,” added Zhu. “We aim to advance embodied intelligence by combining high-fidelity physical sensing with intelligent control systems, contributing to the development of truly general artificial intelligence.”
Written for you by our author Ingrid Fadelli, edited by Sadie Harley, and fact-checked and reviewed by Robert Egan—this article is the result of careful human work. We rely on readers like you to keep independent science journalism alive. If this reporting matters to you, please consider a donation (especially monthly). You’ll get an ad-free account as a thank-you.
More information:
Zihang Zhao et al, Embedding high-resolution touch across robotic hands enables adaptive human-like grasping, Nature Machine Intelligence (2025). DOI: 10.1038/s42256-025-01053-3
© 2025 Science X Network
Citation:
Bio-mimetic robotic hand seamlessly integrates tactile feedback to outperform predecessors (2025, June 11)
retrieved 11 June 2025
from https://techxplore.com/news/2025-06-bio-mimetic-robotic-seamlessly-tactile.html
This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no
part may be reproduced without the written permission. The content is provided for information purposes only.
