Researchers from MIT's Computer Science and Artificial Intelligence Laboratory (CSAIL) have developed an adaptive smart glove that can capture, reproduce, and relay touch-based instructions to teach physical skills. Published in an open-access paper in Nature, the research details a glove that uses digital embroidery to integrate tactile sensors and haptic actuators into the fabric. It can send vibrational feedback to different parts of the hand, guiding the user on optimal hand motions for various tasks.
In one demonstration, an expert pianist recorded a simple tune while wearing the glove. The sensors captured the sequence of key presses. This was converted into personalized haptic instructions fed to a student's glove, directing their fingers to the right keys.
The researchers customize each glove's fit and sensor/actuator layout for the user's hand size and intended application. Using the embroidery machine, fabrication takes under 10 minutes to stitch the sensors and haptics in. An adaptive machine learning model is used to optimize the feedback for each user's unique perception of touch and personalizes the touch feedback with just 15 seconds of data from them.
Experiments showed the glove improving performance in piano playing, time-sensitive gaming, and robotic teleoperation. The haptic directions helped users achieve higher scores in rhythm and racing videogames. In robotic grasping tests, the glove allowed better control of grip pressure, enabling more delicate grasps without deforming soft objects.
The applications of this technology extend far beyond musical instruction. In robotics, for instance, the glove can transfer force sensations from a robot to a human operator, enhancing teleoperation and enabling more delicate manipulations. This feature could revolutionize manufacturing environments, allowing humans and robots to collaborate more safely and efficiently. Furthermore, in virtual and augmented reality settings, the glove introduces a tactile dimension, creating more immersive experiences that could revolutionize training for surgeons, pilots, and other professionals requiring high precision