Rutgers researchers developed a prototype smart shoe that analyzes walking patterns without needing a rechargeable battery.
The shoe uses artificial intelligence to classify movements, including walking at different speeds, running, and climbing stairs.
The technology achieved 95.4% accuracy in testing, but more research is needed before it can be used to monitor patients with movement disorders.
Smartwatches and fitness trackers can help people keep tabs on their activity, but Rutgers University engineers are exploring another way to monitor movement: putting the technology directly into a shoe.
The researchers developed a prototype sneaker that analyzes how people walk and uses energy generated by their own footsteps to power its electronics. The design could eventually help doctors monitor changes in walking patterns among people with Parkinson’s disease, spinal cord injuries, traumatic brain injuries, and other conditions that affect movement.
Walking patterns, also known as gait, include details such as speed, balance, stride, and rhythm. Changes in these patterns can offer clues about a person's health, risk of falling, or progress during rehabilitation.
Currently, doctors often assess gait by watching patients walk during relatively short appointments. A wearable device that collects information over longer periods could eventually provide a more complete picture of how someone moves in everyday life.
How researchers developed and tested the shoe
The team, led by Rutgers biomedical engineering professor Simiao Niu, designed a system that combines movement sensors, a small processor, and artificial intelligence. The findings were published in the journal Science Advances.
The shoe contains a device in its sole that generates electricity from the pressure and friction created with each step. A specially designed circuit converts that energy into a usable form, allowing the electronics to operate without a conventional rechargeable battery.
“When you are walking or running, you automatically have biomechanical energy available, so you can harvest this energy,” Niu said in a news release.
“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. Once you put it onto the charger, you typically forget about it, and then you don’t wear it.”
An accelerometer measures foot movement along three axes, while an AI algorithm analyzes the measurements. The system classifies activity in 15-second intervals, identifying slow walking, fast walking, running, or climbing stairs. The results appear on a screen attached to the shoe.
Researchers initially developed an AI model that examined 21 movement characteristics. However, it required more memory than the shoe's processor could accommodate. By focusing on variations in movement along the three axes, they created a smaller model that ran faster and required less power.
The researchers developed and evaluated the system using data from four healthy volunteers between the ages of 23 and 26.
What the findings could mean for consumers
The streamlined AI model classified movements with 95.4% accuracy. It also ran about 15 times faster and used approximately one-sixth as much current as the original model. Laboratory testing showed that even slow walking generated enough electricity to power the system.
If further research supports its use, the technology could eventually help healthcare providers track changes in patients' movement, assess rehabilitation progress, or identify unusual walking patterns that may warrant closer attention.
The prototype is still in its early stages. It has not been tested in older adults, people with movement disorders, or patients undergoing rehabilitation. It also cannot currently diagnose medical conditions, predict falls, or determine whether a treatment is working.
The findings offer an early look at how everyday footwear might eventually help people and their healthcare providers understand changes in movement without adding another device to charge.