Researchers from the Wyss Institute and Harvard University have developed an algorithm than allows wearable robots to adapt to an individual’s movements in as little as 20 minutes – greatly increasing walking efficiency.
This points to an exciting future of assisted movement across a variety of new applications.
We all move differently, and when we walk, we’re constantly adjusting how we move to save energy – or more accurately, to reduce the metabolic cost.
For training athletes, fitness fanatics, patients who needs movement assistance, or anyone who may be recuperating from injury or illness, soft, assistive devices, like the exoskeleton being developed by Harvard Biodesign Lab can aid these movements by sensitively augmenting the wearer’s physiology, providing the right level of assistance, at the right time.
However, they need to be tailored to the wearer to suit their individual movements. For all the advanced material and robotics engineering that goes into such devices, personalising them for different wearers’ gaits is time-consuming and inefficient.
Gait keepers
Joint research by the Wyss Institute for Biologically Inspired Engineering and the Harvard John A. Paulson School of Engineering and Applied and Sciences (SEAS) has created a machine learning algorithm that can quickly understand an individual’s characteristic movements and tailor the control strategies of soft, wearable exosuits to match.
In a Wyss Institute report, Ye Ding, a Postdoctoral Fellow at SEAS and co-lead author of the research, said:
This new method is an effective and fast way to optimise control parameter settings for assistive wearable devices. Using this method, we achieved a huge improvement in metabolic performance for the wearers of a hip extension assistive device.
The solution is known as a human-in-the-loop Bayesian optimisation method. It helps reduce the metabolic cost of the wearer when compared to walking without the device, or an un-optimised version, by providing personalised hip-assistance. Watch the video, below, for more details.
The algorithm quickly identifies the best control parameters for an individual – to minimise the energy required for walking – by measuring physiological signals, such as breathing rate, to identify the metabolic cost. As the video demonstrates, the system fine-tunes these parameters and adapts the exosuit to the wearer’s needs.
The benefits of an AI-powered exosuit
“Before, if you had three different users walking with assistive devices, you would need three different assistance strategies,” said Myunghee Kim, Ph.D., postdoctoral research fellow at SEAS and co-lead author of the paper.
As well as the time-saving advantages, the combination of algorithm and exosuit reduced metabolic cost by over 17 percent, an improvement of more than 60 percent over the team’s previous work.
Scott Kuindersma, Ph.D., assistant professor of Engineering and Computer Science at SEAS said:
Optimisation and learning algorithms will have a big impact on future wearable robotic devices designed to assist a range of behaviours. These results show that optimising even very simple controllers can provide a significant, individualised benefit to users while walking. Extending these ideas to consider more expressive control strategies and people with diverse needs and abilities will be an exciting next step.
Internet of Business says
This pioneering research shows the far-reaching value that AI has, even in wearable robotics in which the wearer is essentially in control. The next step will be extending the AI’s capabilities to a more complex exoskeleton, assisting multiple joints at the same time.
Exosuits have huge potential across multiple fields. In healthcare, they could assist the elderly and disabled with their movements. In supply chain, manufacturing, construction and agriculture, they could assist with heavy lifting. Similarly, first responders in emergencies and military personnel could also benefit from robotic aids.
While existing solutions are far from Iron Man levels of advancement, there is also some way to go in making exosuits more practical for prolonged wear. However, there’s no doubt that, at the current rate of progress, we’ll soon be seeing wearable robots outside the research lab.