NRI: FND: Robotic Human Enhancement Enabled through Wearable Hip Exoskeletons Capable of Community Ambulation

NRI：FND：通过可进行社区行走的可穿戴髋关节外骨骼实现机器人人类增强

• 批准号: 1830215
• 负责人: Aaron Young
• 金额: $ 70.24万
• 依托单位: Georgia Tech Research Corporation
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2018
• 资助国家: 美国
• 起止时间: 2018-09-01 至 2023-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1830215&HistoricalAwards=false
• 关键词: NRI; FND; Robotic; Human; Enhancement

The increased metabolic and biomechanical demands of ambulation limit community mobility in persons with lower limb disability due to neurological damage or advanced age. Robotic exoskeletons have the potential to assist these individuals to increase community mobility to improve quality of life. Current technology does not support dynamic movements, such as transitioning between different gaits and supporting a wide variety of walking speeds. This project proposes to meet this challenge by exploring an innovative myoelectric controller that brings together information across multiple muscles to assist the exoskeleton in a range of tasks for mobility. Of interest is how to extract information about the human user's intent, such as what speed the user wants to walk at or if they want to take a step up a stair. Recognizing intent will allow the device to give appropriate assistance over a wide range of activities. For this research project, the investigators will use a new type of robotic hip exoskeleton to augment the human hip during tasks such as walking at different speed, ramps and stairs. This project seeks to advance the state of the science in man-machine exoskeleton interfaces through new types of control techniques. This will help the team's long-term research goal of creating robotic assistive devices that benefit individuals with serious lower limb disabilities by improving mobility and, thus, overall independence. This NSF project will also include a significant outreach to the local Atlanta community. Local project partners will bring underserved minority students in the Atlanta area to Georgia Tech to participate in this educational program. These high school students will interact with the human assistive robotics in the lab and design, build, and program their own assistive robots through hands on education. This project will facilitate the training of an interdisciplinary group of students including graduate and undergraduate biomedical, electrical, and mechanical engineers, GT graduate students training to be clinical practitioners in prosthetics and orthotics, and physical therapists in training in Emory's DPT program. This interdisciplinary group of students will work together to fully integrate all aspects of the project and facilitate learning.This research project will advance the knowledge of myoelectric (EMG) control for enabling humans to have dynamic movement assistance using lower limb robotic exoskeletons. Existing technologies of exoskeleton systems have limited high-level understanding of user intent, thus precluding adaptation to a proper range of daily tasks of living. Sensing modalities used in such systems do not provide sufficient information regarding key physiological parameters such as muscle activity. Conventional control algorithms relate a single modality of sensors to exoskeleton assistance and are thus incapable of fusing broader sets of varying sensor information. These technological gaps have impeded the translation of such systems beyond lab settings to clinical use where they can impact important health needs such as assistive rehabilitation. This proposal will address these gaps in assistive robotic technology by pursuing research organized in three key objectives: Objective 1) Determine the most effective strategy of providing exoskeleton hip assistance for reducing metabolic cost using a novel myoelectric controller. Objective 2) Compare the metabolic and biomechanical effects of a novel controller driven by myoelectric inputs vs a standard controller driven by kinematic inputs. Objective 3) Determine the contributions of high-level intent recognition using myoelectric information to improve control of a powered hip exoskeleton over simulated community terrain. Relatively simple mobility tasks like standing from a seated position or walking represent the primary outcome measures that determine independence in rehabilitation. Thus, the wearable exoskeleton system described in this grant proposal has the potential to make a measurable impact on enhancing the functional performance capabilities of individuals with lower limb deficits by increasing their quality of life, independence and well-being.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.

由于神经损伤或高龄导致的下肢残疾患者，行走的代谢和生物力学需求的增加限制了社区活动能力。机器人外骨骼有可能帮助这些人增加社区流动性，提高生活质量。目前的技术不支持动态运动，例如在不同的步态之间转换和支持各种步行速度。该项目提出通过探索一种创新的肌电控制器来应对这一挑战，该控制器将多个肌肉的信息整合在一起，以帮助外骨骼完成一系列行动任务。人们感兴趣的是如何提取有关人类用户意图的信息，例如用户想要以什么速度行走，或者他们是否想要走上楼梯。识别意图将允许设备在广泛的活动中提供适当的帮助。在这个研究项目中，研究人员将使用一种新型的机器人髋关节外骨骼来增强人类髋关节在不同速度、斜坡和楼梯行走等任务中的功能。该项目旨在通过新型控制技术推动人机外骨骼界面的科学发展。这将有助于该团队的长期研究目标，即创造机器人辅助设备，通过提高移动性和整体独立性，使患有严重下肢残疾的个人受益。这个NSF项目也将包括一个重要的延伸到当地的亚特兰大社区。当地的项目合作伙伴将把亚特兰大地区服务不足的少数民族学生带到佐治亚理工学院参加这个教育项目。这些高中生将在实验室中与人类辅助机器人互动，并通过动手教育设计、建造和编程他们自己的辅助机器人。该项目将促进跨学科学生群体的培训，包括研究生和本科生生物医学，电气和机械工程师，GT研究生培训成为假肢和矫形的临床从业人员，以及Emory DPT项目培训的物理治疗师。这个跨学科的学生小组将共同努力，充分整合项目的各个方面，促进学习。该研究项目将推进肌电（EMG）控制的知识，使人类能够使用下肢机器人外骨骼进行动态运动辅助。现有的外骨骼系统技术对用户意图的理解程度有限，因此无法适应日常生活任务的适当范围。在这种系统中使用的传感模式不能提供关于关键生理参数（如肌肉活动）的足够信息。传统的控制算法将传感器的单一模式与外骨骼辅助联系起来，因此无法融合更广泛的不同传感器信息集。这些技术差距阻碍了这些系统从实验室环境向临床应用的转化，在临床应用中，它们可以影响辅助康复等重要的卫生需求。本提案将通过以下三个关键目标来解决辅助机器人技术的这些空白：目标1)确定使用新型肌电控制器提供外骨骼髋关节辅助以降低代谢成本的最有效策略。目的2)比较由肌电输入驱动的新型控制器与由运动输入驱动的标准控制器的代谢和生物力学效应。目的3)利用肌电信息确定高水平意图识别的贡献，以改善动力髋关节外骨骼在模拟社区地形上的控制。相对简单的活动任务，如从坐姿站立或行走是决定康复独立性的主要结果措施。因此，本拨款提案中描述的可穿戴外骨骼系统有可能通过提高下肢缺陷患者的生活质量、独立性和幸福感，对增强其功能表现能力产生可衡量的影响。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。

项目成果

Real-Time Walk Detection for Robotic Hip Exoskeleton Applications

• DOI: 10.1109/ismr48347.2022.9807510
• 发表时间: 2022-04
• 期刊: 2022 International Symposium on Medical Robotics (ISMR)
• 作者: Hang Man Cho;Inseung Kang;DongHo Park;Dean D. Molinaro;Aaron J. Young

Design and Validation of a Cable-Driven Asymmetric Back Exosuit

• DOI: 10.1109/tro.2021.3112280
• 发表时间: 2021-10-04
• 期刊: IEEE TRANSACTIONS ON ROBOTICS
• 影响因子: 7.8
• 作者: Li, Jared M.;Molinaro, Dean D.;Young, Aaron J.
• 通讯作者: Young, Aaron J.

Wearable Sensor-Based Step Length Estimation During Overground Locomotion Using a Deep Convolutional Neural Network

使用深度卷积神经网络进行地上运动期间基于可穿戴传感器的步长估计

• DOI: 10.1109/embc46164.2021.9630060
• 发表时间: 2021
• 期刊: 2021 43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC
• 作者: Jin, Heejoo;Kang, Inseung;Choi, Gayeon;Molinaro, Dean D.;Young, Aaron J.
• 通讯作者: Young, Aaron J.

Reduction of Trunk Extensor Muscle Activation using a Cable-Driven Asymmetric Back Exosuit

使用电缆驱动的不对称背部外装减少躯干伸肌激活

• 发表时间: 2021
• 期刊: American Society of Biomechanics
• 作者: Molinaro, D;King, A;Mazumdar, A;Young, A
• 通讯作者: Young, A

Subject-Independent, Biological Hip Moment Estimation During Multimodal Overground Ambulation Using Deep Learning

• DOI: 10.1109/tmrb.2022.3144025
• 发表时间: 2022-02-01
• 期刊: IEEE TRANSACTIONS ON MEDICAL ROBOTICS AND BIONICS
• 作者: Molinaro, Dean D.;Kang, Inseung;Young, Aaron J.
• 通讯作者: Young, Aaron J.