Lower-body exoskeletons don't merely thrust the wearer's legs forward – they move them based on the user's existing gait. A team at Harvard University has now developed what is claimed to be a better method of gauging that gait, utilizing ultrasound.
Ordinarily, exoskeletons are calibrated to individual users' stride length, pace, pronation and other gait characteristics simply by detecting and analyzing actual leg movements. This process may require the person to walk around for several hours while wearing the apparatus – it's a task which might not even be possible for users with seriously limited mobility.
Seeking to speed up and simplify the process, the Harvard researchers instead utilized portable ultrasound transducers to image and record the calf muscles of volunteers as they performed various walking tasks on a treadmill. Based on the resulting muscle-activity profiles, the team proceeded to estimate how much assistive force should be applied by a lower-leg exoskeleton during the push-off phase of each person's walking cycle.
The individual profiles could be generated via just a few seconds of walking. And after those profiles were programmed into the exoskeleton, the test subjects were found to use less metabolic energy when walking with the exoskeleton than when without it.
Additionally, though, the exoskeleton was able to use less force to deliver that level of assistance, as compared to when the ultrasound calibration technique wasn't used. This means that the device was not only gentler on its users, but it was also more energy efficient.
"This study shows that you can provide more effective walking assistance if you time it to when the muscle starts contracting, rather than starting the assistance based on how the leg is moving," study co-author Prof. Robert Howe tells us. "It turns out there’s considerable variation between individuals in the timing of calf muscle contraction, and the ultrasound lets you determine the best time for assistance for each individual."
The research is described in a paper that was recently published in the journal Science Robotics.
Source: Harvard John A. Paulson School of Engineering and Applied Sciences
