New wearable AI device could help blind people navigate independently

Overview of the wearable multimodal visual assistance system. Credit: Nature Machine Intelligence (2025).

A new wearable device powered by artificial intelligence (AI) is making it easier for blind and partially sighted people to move around safely.

The system, described in the journal Nature Machine Intelligence, helps users detect obstacles and find clear paths using sound and gentle vibrations.

For people who are blind or have low vision, getting around can be challenging, especially in unfamiliar places.

While some medical treatments and surgical implants can help, they’re not always effective or accessible.

Wearable devices offer a non-invasive option, but most current technologies are hard to use, which limits how many people actually adopt them.

That’s why researcher Leilei Gu and their team created a more user-friendly wearable system that uses AI to “see” the environment and guide the user in real time.

The device includes a camera that scans the surroundings and sends video to an AI algorithm. The algorithm processes the video to identify obstacles and find the best path forward.

Once the system maps a safe route, it gives directions through two different methods.

First, bone conduction headphones send audio instructions directly to the user’s inner ear, allowing them to hear guidance clearly while still being aware of outside sounds.

Second, special wearable patches—called artificial skins—are worn on the wrists. These skins gently vibrate to let the user know if there’s something on either side that they should avoid.

To test the device, the researchers used both robot simulations and real-life trials with blind and partially sighted volunteers. The participants used the system to move through virtual mazes and real-world environments. They showed noticeable improvement in their ability to avoid obstacles and perform tasks like reaching for and picking up objects.

The results suggest that combining visual, audio, and touch signals can make assistive devices much more effective and easier to use. The mix of these three senses helps users understand their environment in a more natural and intuitive way.

The researchers hope that with further improvements, this AI-powered device can become a practical and widely used tool—not only for personal navigation but also for other areas of assistive technology. This breakthrough could help millions of people with vision loss gain more independence and confidence in their daily lives.