A research team led by Professor Im-Doo Jeong of the Department of Mechanical Engineering at UNIST has developed a smart contact lens capable of controlling a robotic arm using only eye movements. This is significant as a human-machine interface that converts eye movements directly into machine control signals, without the need for head-mounted devices or handheld controllers.

According to UNIST, the results of this study were published in the materials science journal Advanced Functional Materials in March 2026. The title of the paper is 'Meniscus Pixel Printing for Contact-Lens Vision Sensing and Robotic Control'.

This smart contact lens works by tracking the direction of gaze by reading the light distribution that changes according to eye movement through 100 light sensors arranged in a 10×10 array integrated on the lens. It is designed to distinguish not only up, down, left, and right, but also diagonals, and recognizes blinking as a separate command so that the robotic arm can even perform the action of picking up an object.

The research team applied 'meniscus pixel printing (MPP)' technology to directly implement a sensor on the curved surface of a lens. By enabling the formation of sensor patterns on curved surfaces without the need for masks or complex multi-step processes, the focus was on reducing the limitations of existing flat-plane-centered fabrication methods.

The low signal resolution caused by the narrow lens area was compensated for by artificial intelligence. Although there are 100 actual sensors, a deep learning-based super-resolution reconstruction model reconstructed a 10×10 input into 80×80 level optical information, and the latency was presented as 0.03 seconds.

In experiments using an eye model, the robot arm was able to pick up and move an object using only eye movements. The research team stated that the accuracy of recognizing gaze gestures in nine directions was 99.3%. This result is interpreted as demonstrating that control performance can be secured even within a small device like a lens by combining sensor manufacturing and AI signal restoration technology.

The first authors of this study are researchers Gong Byeong-hun and Kim Do-hyeon, and Professor Jeong Im-doo is listed as a co-author. The research was conducted with government support, including from the National Research Foundation of Korea.