Stanford University researchers have built an exoskeleton that employs machine learning to adjust to its wearers’ stride, enabling people with restricted mobility to walk more easily.
The exoskeleton, which resembles a mechanized boot, is lightweight and allows the wearer to move somewhat freely, increasing walking speed while decreasing energy consumption.
It is made of inexpensive wearable sensors, a motor, and a small Raspberry Pi computer, all powered by a rechargeable battery pack worn around the waist. The sensors are discreetly implanted in the boot to measure force and motion.
The boot’s data is sent into a machine-learning model, tuning the device to tailor its support, applying force at the ankle to replace some of the calf muscle’s function, and assisting the wearer in pushing off the ground while taking a step. This allows you to walk faster and with less effort. It just takes an hour for the model to personalize how the device supports the wearer, and because it is constantly learning from sensor data, the device may adapt over time as the wearer’s walk changes.
Compared to walking in standard shoes, the device resulted in a 9% increase in walking speed and a 17% reduction in energy consumed while walking naturally.
Compared to other devices on a treadmill, the exoskeleton produced around double the effort decrease. In addition, the energy savings and speed gain were compared to “taking off a 30-pound backpack,” according to the study.
Although supportive exoskeletons have been available for a while, they are difficult to adjust to the particular wearer due to their size and weight. As a result, their success has mostly been limited to treadmills within laboratories, which are expensive. The Stanford team’s exoskeleton is far smaller than others on the market and, more importantly, much easier to use.
According to the researchers, this is the first time an exoskeleton has shown the ability to conserve human energy in real-world situations. They believe it will allow older persons with limited mobility or people with muscular difficulties to walk around more easily.
According to Kaspar Althoefer, head of the Center for Advanced Robotics at the Queen Mary University of London, who was not engaged in the study, the team should be able to bring the device to a large market eventually.
“It would be beneficial for those who are maybe not so strong anymore and want to walk a bit further,” he says.
The findings of the study are published in Nature.