video: A brief, one-minute video describing how Zhang et al. personalize an exoskeleton for its user, decreasing the amount of effort required to use it. This material relates to a paper that appeared in the 23 June 2017, issue of Science, published by AAAS. The paper, by J. Zhang at Carnegie Mellon University in Pittsburgh, PA, and colleagues was titled, "Human-in-the-loop optimization of exoskeleton assistance during walking." view more
Credit: Rahkendra Ice / AAAS
Researchers have developed an exoskeleton system that provides personalized support for its user. In healthy volunteers, the optimized exoskeleton reduced energy expenditure during walking by 24%, on average, compared to when the system was not providing personalized support. Exoskeletons can help enhance human abilities, for example by allowing people to lift greater weight or expend less energy during rehabilitation. Yet, because each person exhibits unique natural movements, a one-size-fits-all exoskeleton approach will not work, research shows. For example, one person may rotate their ankle more than another, as they walk. To create a more personalized exoskeleton, Juanjuan Zhang et al. developed a system that measures an individual's energy expenditure as the exoskeleton, which supports the user's ankle during walking, subtly changes its pattern of assistance. In this case, the system alters its support of ankle movements in four areas: peak torque (rotation), timing of peak torque, and rise and fall times. Over the course of about one hour of walking, the system identifies which patterns of support help the person expend the least amount of energy. With the exoskeleton optimized for torque support on just one ankle, 11 volunteers experienced a 24% reduction in energy expenditure, on average, while walking. One volunteer who wore the optimized exoskeleton on both ankles experienced a 33% reduction in energy expenditure. Intriguingly, participants who engaged in additional rounds of walking with the optimized exoskeleton experienced further reductions in energy expenditure, suggesting that users who stick with the device may undergo their own subtle adaptations, resulting in further optimization. This advance is highlighted in a Perspective by Philippe Malcolm et al.
###
Journal
Science