These Smart Glasses Don’t Need Charging—they Draw Power From Your Face

Researchers at the McCormick School of Engineering at Northwestern University have developed a new type of smart glasses equipped with miniature sensors that can harvest mechanical energy from facial movements and convert it into electrical energy. Even a quick blink or the subtle pulsation of an artery provides enough mechanical energy to activate these highly sensitive sensors.

To test these self-powered sensors, the researchers used the glasses to detect actions such as clenching the jaw, tapping the teeth together, blinking, and raising the eyebrows. The glasses were not only able to distinguish these specific movements from everyday activities like talking or eating, but also achieved a 99% accuracy rate in identifying teeth-tapping and a 96% accuracy rate in identifying jaw-clenching.

This development holds the potential to enable ordinary eyeglasses to seamlessly monitor cardiovascular health, bruxism (involuntary teeth clenching and grinding), fatigue, eating behaviors, and other daily activities—all without the need for bulky batteries or regular recharging.

At the UbiComp/ISWC 2026 conference, researchers presented this innovation, known as GlassTENG.

A little static electricity can be put to great use

At the heart of these new glasses are three paper-thin devices known as triboelectric nanogenerators (TENGs). While the name may sound complex, the underlying principle is quite familiar: static electricity.

Anyone who has rubbed a balloon against their hair on a dry winter day—or petted a cat—has experienced the triboelectric effect. When two different materials come into contact and then separate, electrons shift from one material to the other, generating a momentary electrical charge.

GlassTENG leverages this fundamental principle.

Each flexible sensor consists of multiple layers of material that are repeatedly pressed together and pulled apart by facial movements. For instance, blinking compresses one sensor; clenching one’s teeth activates another; and even the subtle pressure from blood flowing through an artery is enough to trigger a sensor.

Three sensors, multiple signals

To capture as much information as possible, the team placed three sensors at key locations around the eyeglass frame. One is positioned at the bridge of the nose—near the angular artery and the muscles surrounding the eyes—to detect pulses, as well as blinking and eyebrow movements. Another is located near the temple, where it monitors the superficial temporal artery and jaw movements. A third is placed near the ear to detect actions such as chewing and clenching the teeth.

Together, these three sensors provide the glasses with different perspectives for "sensing" facial movements. The sensors are sensitive to a range of forces: arterial pulses—the lightest force tested by the researchers—generate electrical signals of approximately 50 to 100 millivolts. Blinking and eyebrow movements produce signals of about 200 to 500 millivolts, while more forceful jaw movements, such as chewing or clenching the teeth, can generate signals as high as 1.5 volts.

While wearing GlassTENG, 20 participants performed six activities: blinking, raising their eyebrows, clenching their teeth, tapping their teeth together, eating, and speaking. The research team then used machine learning algorithms to analyze patterns in the electrical signals to determine which activity each participant had performed.

The system achieved an accuracy rate of 93.8% across the six activities.

Ultra-low power consumption, a huge advantage

Beyond its functionality, another major advantage of GlassTENG is that it requires no external power source. Many existing wearable devices consume continuous power, leading to higher energy demands that necessitate larger batteries—which in turn increase both weight and bulk. Furthermore, the need for regular battery recharging poses a significant obstacle to truly continuous health monitoring.

The three sensors in GlassTENG consume a combined total of only 4.1 microwatts of power—several orders of magnitude less than many conventional sensing technologies. The device then transmits data directly to a smartphone or tablet via Bluetooth.

Source: VRAR Planet