en.Wedoany.com Reported - Researchers at the Massachusetts Institute of Technology (MIT) have recently developed an ultrasonic wristband designed to collect data on human hand movements, aiming to help humanoid robots master dexterous tasks such as grasping and manipulating objects.

Although artificial intelligence excels at handling computer-based tasks, humanoid robots still face challenges when performing tasks requiring fine manipulation in the physical world, such as grasping a cup. This wristband offers a new approach to solving this problem. It uses high-frequency sound waves to "see through" the wearer's skin, capturing movements of muscles, tendons, and ligaments beneath the surface. The images are then transmitted to a computer, where AI algorithms decode them into specific gestures.

Xuanhe Zhao, a professor of mechanical engineering at MIT, explained that this technology can be used to collect data on household chores, training robots to perform these tasks with dexterous hand movements. Additionally, the technology holds promise for applications in other fields requiring fine hand movements, such as surgical procedures.

The images generated by the wristband are processed by AI algorithms to parse what engineers call "degrees of freedom"—the specific ways joints bend or rotate. The human hand has a total of 22 degrees of freedom, and even tracking a small subset of these movements posed significant challenges in early systems.

In a laboratory demonstration involving eight volunteers, the wristband demonstrated the ability to accurately mirror gestures within 120 milliseconds, recognizing all 26 letters of the American Sign Language alphabet. Additionally, the wristband supports wireless operation, meaning the controller and the receiving robot do not need to be in the same physical space.

Beyond remote control, the research team also plans to use the wristband to build a large dataset of human movements, with the goal of enabling humanoid robots to eventually learn dexterous manipulation tasks autonomously, without human guidance.

This article is compiled by Wedoany. All AI citations must indicate the source as "Wedoany". If there is any infringement or other issues, please notify us promptly, and we will modify or delete it accordingly. Email: news@wedoany.com