An exosuit tailored to fit

The algorithm allows personalized control strategies for soft, wearable exosuits.

Harvard researchers have developed an efficient machine-learning algorithm that can quickly tailor personalized control strategies for soft, wearable exosuits, significantly improving the devices’ performance.
Harvard researchers have developed an efficient machine-learning algorithm that can quickly tailor personalized control strategies for soft, wearable exosuits, significantly improving the devices’ performance.

Wearing exosuit requires human or robot to be in sync. But every human moves a bit differently, and tailoring the robot’s parameters to an individual user is a time-consuming and inefficient process. Thus, Harvard scientists designed have developed an efficient machine-learning algorithm that can do that work quickly.

The new algorithm cuts through that variability to rapidly identify the best control parameters to minimize the work of walking. It uses human-in-the-loop optimization, which uses real-time measurements of human physiological signals, such as breathing rate, to adjust the control parameters of the device. Whenever a user wears exosuit, the algorithm old the exosuit when and where to deliver its assistive force to improve hip extension.

Ye Ding, a postdoctoral fellow at SEAS and co-first author of the research said, “This new method is an effective and fast way to optimize 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.”

Myunghee Kim, a postdoctoral research fellow at SEAS said, “Before, if you had three different users walking with assistive devices, you would need three different assistance strategies. Finding the right control parameters for each wearer used to be a difficult, step-by-step process.”

The mix of the algorithm and suit, diminished metabolic cost by 17.4 percent contrasted and strolling without the device, a change of in excess of 60 percent over the group’s past work.

Conor Walsh, the John L. Loeb Associate Professor of Engineering and Applied Sciences said, “With wearable robots like soft exosuits, it is critical that the right assistance is delivered at the right time so that they can work synergistically with the wearer. With these online optimization algorithms, systems can learn how to achieve this automatically in about 20 minutes, thus maximizing benefit to the wearer.”

Now, scientists are further planning to apply the optimization to a more complex device that assists multiple joints, such as hip and ankle, at the same time.