Site iconSite icon ForkLog

Stanford researchers develop a lightweight exoskeleton with artificial intelligence

uskoryayushhei-sya-fragmentatsii-mirovogo-interneta

Researchers from Stanford University have developed an exoskeleton with sensors and artificial intelligence algorithms designed to enhance its performance. Technology Review reports.

The robotic device resembles a boot. It is equipped with inexpensive wearable sensors, a motor, and a Raspberry Pi single-board computer powered by a battery.

The exoskeleton, equipped with sensors and artificial intelligence algorithms. Data: Stanford University.

The data generated by the boot are analyzed by a machine-learning model. It tunes the device and moves the ankle, replacing some functions of the calf muscle. As a result, the leg pushes off the ground and takes a step, allowing the wearer to walk faster with less effort.

According to the developers, the AI model is personalised in just one hour. In operation, the algorithm continues to train, adapting to changing user behaviour.

In tests, the researchers found that use of the device increased walking speed by 9% and cut energy expenditure by 17% versus natural movement in ordinary footwear.

On a treadmill, the device was twice as effective as similar devices. The researchers compared energy savings and speed increases to the \”removal of a 14-kilogram backpack\”.

According to the developers, existing exoskeletons are difficult to adapt to an individual user due to their bulkiness. They are also expensive and rarely show positive results outside laboratories.

Stanford highlighted the size and weight of their device. They say the device marks the first demonstration of an exoskeleton’s ability to save human energy in real-world conditions.

Researchers hope the development will enable older people or those with muscular problems to move more freely.

\”This would be very helpful for those who may no longer be as strong and want to go a little further,\” said Caspar Altofer, head of the Centre for Advanced Robotics at Queen Mary University of London, who was not involved in the study.

In the future, the researchers plan to test the device on elderly people and commercialise the product.

\”We hope that it will be somewhat like an electric bicycle. It doesn’t do all the work for you, but requires a reasonable amount of effort,\” said one of the project authors, Patrick Slade.

As reported in January 2022, an AI-powered eye implant restored the vision of an 88-year-old woman.

In March 2021, researchers demonstrated an AI backpack to assist the visually impaired in orienting themselves.

Follow ForkLog’s news on Telegram: ForkLog AI — all the news from the world of AI!

Exit mobile version