Collaborative Research: Towards Task-Agnostic and Device-Agnostic Ankle Exoskeleton Control for Mobility Enhancement

外骨骼在改善或增强人体活动能力方面具有巨大的潜力。然而，尽管最近的研究显示了对行走等简单任务的影响，但研究人员往往忽略了日常生活中必要的更复杂的动作，比如在厨房里拖着脚走，弯腰拿起背包，或者进出车辆。因此，外骨骼目前无法在高度有限的任务范围内提供帮助，甚至可能降低这些更复杂任务的性能。此外，它们必须在特定主题的基础上进行调整，以获得足够的性能，这阻碍了广泛采用，并使许多人无法使用这些设备。该项目旨在通过利用可穿戴传感和机器学习的最新进展，彻底改变下肢外骨骼领域。拟议的研究包括将传感器读数映射到外骨骼辅助，并需要收集外骨骼佩戴者执行各种任务的运动数据。目标是使外骨骼能够协助所有可能的日常生活任务，增加其实用性和可及性，为更广泛的人。这与NSF推进国民健康和福利的使命是一致的，因为这项新技术将显著提高未来的行动能力和独立性，特别是在老年人和神经肌肉缺陷患者中。该项目将开放其数据集和深度学习模型，通过使其他研究人员能够在我们的进步基础上进一步推动该领域的技术创新，从而促进科学的进步。最后，该项目通过与生物力学和机器人相关的外展活动支持教育和多样性，将少数民族高中的学生带到东北大学和佐治亚理工学院，进行可穿戴机器人的动手演示和比赛。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Exoskeletons have tremendous potential to improve or augment human mobility. However, despite recent work showing the impact on easily defined tasks such as walking, researchers have tended to ignore the more complex maneuvers that are necessary in daily life, such as shuffling in the kitchen, bending to pick up a backpack, or transitioning in and out of a vehicle. Accordingly, exoskeletons are currently incapable of assisting beyond a highly limited range of tasks and may even reduce performance in these more complex tasks. Further, they must be tuned on a subject-specific basis for adequate performance, hindering broad adoption and leaving the devices inaccessible for many people. This project aims to revolutionize the field of lower limb exoskeletons by leveraging recent advancements in wearable sensing and machine learning. Proposed research involves mapping sensor readings to exoskeleton assistance and requires collecting motion data exoskeleton wearers performing a wide range of tasks. The goal is to make exoskeletons capable of assisting with all possible tasks of daily living, increasing their utility and accessibility for a broader range of people. This is aligned with NSF's mission to advance national health and welfare, as this new technology will significantly improve mobility and independence in the future, particularly during aging and for individuals with neuromuscular deficits. This project will open-source its datasets and deep learning models, promoting the progress of science by enabling other researchers to build upon our advances and further drive technological innovation in this domain. Finally, this project supports education and diversity through outreach events related to biomechanics and robotics, bringing students from minority-serving high schools to Northeastern University and Georgia Tech for hands-on demos and competitions with wearable robotics.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.