You head down the sidewalk, pick up the pace, then take the stairs. Your shoes absorb the pounding while you think about everything except your feet. But those ordinary movements contain information about how your body is working — and enough energy to power a device that helps interpret them.
Engineers have developed a prototype smart shoe that uses footsteps to run artificial intelligence, without a battery to recharge. Described in Science Advances, the system recognizes different activities, counts steps, and estimates calories burned. Eventually, the approach could help clinicians follow changes in how people walk.
“When you are walking or running, you automatically have biomechanical energy available, so you can harvest this energy,” said lead researcher Simiao Niu in a press release.
How a Smart Shoe Turns Footsteps Into Electricity
For Niu and his research team, a shoe offered the convenient combination of access to both movement data and the energy needed to analyze it.
Simiao Niu (right) and Fuying Dong with their footstep-powered smart shoe.
(Image Credit: Veronica Mendez/Rutgers University)
Inside the sole, a device generates electricity through the pressure and friction of walking. It relies on the triboelectric effect, the phenomenon behind the static electricity that builds up when certain materials rub together.
Turning footsteps into a dependable power supply, however, takes more than collecting that charge. Each step delivers an irregular burst of energy that the electronics cannot use directly.
Explore All OptionsThe team developed a power-management circuit to make that electricity usable, increasing the available energy by as much as 120 times compared with a conventional approach. In laboratory tests, even slow walking supplied enough electricity to keep the complete system working.
Read More: First-Ever Single Brain Implant Can Translate Speech and Gestures at the Same Time
Onboard AI Identifies Walking, Running, and Stairs
The shoe tracks foot movement with an accelerometer that measures motion along three axes. A small processor then analyzes 15-second stretches of data, identifying slow walking, fast walking, running, or stair climbing.
The prototype achieved 95.4 percent accuracy at classifying those activities and shows its results on a shoe-mounted screen.
Keeping the analysis inside the shoe also helps stretch its modest energy supply. This approach, called “edge AI,” reduces the need to continually transmit raw measurements to a phone, computer, or remote server. Together, the motion sensor and AI algorithm consume just 86 microwatts.
“We want to solve the fundamental bottleneck in current wearable devices. We are developing a smart wearable with integrated AI functionality that can harvest energy on its own, so you don’t need to worry about charging,” explained Niu.
Why Walking Patterns Could Help Monitor Health
A person’s gait encompasses the rhythm, speed, balance, and stride of their walk. Changes can provide clues about recovery from injury, disease progression, or the likelihood of falling.
“Gait is one of the most significant biomarkers for a lot of diseases,” said Niu. “If you are able to use what I call the ‘worry-free shoes’ we’ve developed, patients can just wear them, and the shoes can automatically collect their gait pattern.”
Clinical assessments often capture only a short walk in a controlled setting. Wearable monitoring could offer a longer view of movement during everyday life, potentially helping people with Parkinson’s disease, traumatic brain injuries, or spinal cord injuries.
Those applications still require further development and clinical testing. For now, the advance is a demonstration that footsteps can power their own analysis. If that approach proves useful in patients, a routine walk could become a way to track recovery or spot changes between appointments.
Read More: Ultramarathon Running Could Rapidly Age and Breakdown Red Blood Cells
Article Sources
Our writers at Discovermagazine.com use peer-reviewed studies and high-quality sources for our articles, and our editors review for scientific accuracy and editorial standards. Review the sources used below for this article:
- This article references information from Science Advances: A biomimetic, ultralow-power edge-AI-empowered and self-sustaining gait analysis system
