Engineers build ‘robotic boot’ that makes walking easier (BINTEO)
Which even has a machine learning algorithm
Engineers and researchers from Stanford University in California, USA, have created a small exoskeleton, which looks like a “robotic boot” and can make walking easier for people with limited mobility.
Going into details, the exoskeleton in question is lightweight and allows the user to move relatively freely, increasing their walking speed while simultaneously reducing the energy consumed for movement. As for the hardware, it consists of cheap wearable sensors, a motor and a Raspberry Pi. The entire system is powered by a rechargeable battery worn around the user’s waist.
The remarkable thing about the case is that the exoskeleton also uses machine learning to adapt to the gait of each person. Thus, the data generated by the boot is fed into a specially configured machine learning model, which in turn tunes the device, changing the force applied to the ankle, for example. In this way, the exoskeleton replaces some of the function of the calf muscle and assists the wearer by providing an extra boost as they lift their foot off the ground as they take a step. This makes it possible to walk faster, with less effort.
The exoskeleton takes just an hour to “learn” the user, and because it’s constantly getting data from sensors, it can further adapt over time as the user’s gait potentially changes.
Overall, the team found that the device led to a 9% increase in walking speed and a 17% reduction in energy expended during natural walking, compared to walking in regular shoes. The researchers likened the energy savings and speed increase to “taking a 30kg backpack off your back”.
The research has the official title ‘Personalizing exoskeleton assistance while walking in the real world’ and was published in the scientific journal nature on October 12, 2022.
You can find it by clicking here.
