John A. Paulson School of Engineering and Applied Sciences, Harvard University, Boston, 02134, USA.
School of Life Sciences, Swiss Federal Institute of Technology (EPFL), Lausanne, Switzerland.
Sci Rep. 2022 Jun 29;12(1):11004. doi: 10.1038/s41598-022-14784-9.
As we age, humans see natural decreases in muscle force and power which leads to a slower, less efficient gait. Improving mobility for both healthy individuals and those with muscle impairments/weakness has been a goal for exoskeleton designers for decades. In this work, we discover that significant reductions in the energy cost required for walking can be achieved with almost 50% less mechanical power compared to the state of the art. This was achieved by leveraging human-in-the-loop optimization to understand the importance of individualized assistance for hip flexion, a relatively unexplored joint motion. Specifically, we show that a tethered hip flexion exosuit can reduce the metabolic rate of walking by up to 15.2 ± 2.6%, compared to locomotion with assistance turned off (equivalent to 14.8% reduction compared to not wearing the exosuit). This large metabolic reduction was achieved with surprisingly low assistance magnitudes (average of 89 N, ~ 24% of normal hip flexion torque). Furthermore, the ratio of metabolic reduction to the positive exosuit power delivered was 1.8 times higher than ratios previously found for hip extension and ankle plantarflexion. These findings motivated the design of a lightweight (2.31 kg) and portable hip flexion assisting exosuit, that demonstrated a 7.2 ± 2.9% metabolic reduction compared to walking without the exosuit. The high ratio of metabolic reduction to exosuit power measured in this study supports previous simulation findings and provides compelling evidence that hip flexion may be an efficient joint motion to target when considering how to create practical and lightweight wearable robots to support improved mobility.
随着年龄的增长，人类肌肉力量和力量自然会下降，导致步态变慢、效率降低。几十年来，提高健康个体和肌肉损伤/虚弱个体的活动能力一直是外骨骼设计师的目标。在这项工作中，我们发现与最先进的技术相比，通过利用人机交互优化来理解个性化髋关节屈曲辅助的重要性，可以将行走所需的能量成本降低近 50%。这是相对未探索的关节运动。具体来说，我们表明，与辅助关闭时的运动相比，系绳髋关节屈曲外骨骼可以将代谢率降低 15.2%±2.6%（相当于与不穿外骨骼相比降低 14.8%）。这种大的代谢减少是通过非常低的辅助幅度（平均 89N，约为正常髋关节屈曲扭矩的 24%）实现的。此外，代谢减少与正外骨骼功率的比值比以前发现的髋关节伸展和踝关节跖屈的比值高 1.8 倍。这些发现促使设计了一种重量轻（2.31kg）且便携式的髋关节屈曲辅助外骨骼，与不穿外骨骼相比，它可将代谢降低 7.2%±2.9%。本研究中测量的代谢减少与外骨骼功率的高比值支持了先前的模拟发现，并提供了有力的证据，表明在考虑如何创建实用且重量轻的可穿戴机器人以支持改善的移动性时，髋关节屈曲可能是一个有效的关节运动目标。
