CPS: Medium: Collaborative Research: User and Environment Interactive Planning and Control of Artificial Lower Limbs for Resilient Locomotion

CPS：中：协作研究：用于弹性运动的人工下肢的用户和环境交互式规划和控制

• 批准号: 10267031
• 负责人: Hartmut Geyer
• 金额: $ 36.49万
• 依托单位: CARNEGIE-MELLON UNIVERSITY
• 依托单位国家: 美国
• 财政年份: 2019
• 资助国家: 美国
• 起止时间: 2019-09-01 至 2023-08-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/10267031
• 关键词: Acceleration; Address; Amputees; Behavior; Biological Sciences; Biomedical Technology; Communities; Complex; Data; Development; Elderly; Engineering; Environment; Foundations; Gait; Goals; Health; Human; Intelligence; Learning; Limb Prosthesis; Locomotion; Lower Extremity; Mission; National Institute of Biomedical Imaging and Bioengineering; Population; Prosthesis; Quality of life; Rehabilitation therapy; Research; Robot; Self-Help Devices; Sensory; Spinal cord injury patients; Stroke; System; Target Populations; Technology; Time; collaborative environment; cyber physical; design; exoskeleton; experimental study; gait examination; gait rehabilitation; human subject; improved; innovation; mobility aid; mobility rehabilitation; multisensory; next generation; physical science; prototype; robot exoskeleton; sensor

Today, even normal locomotion is difficult to master for people who depend on prostheses or exoskeletons for mobility and rehabilitation, and there is a clear understanding in the wearable robotics community that a more interactive control of artificial limbs will be required than is common today to overcome this limitation. This project embraces such an interactive control. A cyber-physical approach to control is pursued, in which the artificial limb takes advantage of rich sensory information to continuously reason about and adapt its behavior to both the user and the environment in a way that improves locomotion stability, robustness and versatility. The specific aims of the proposed research are to establish this cyber-physical approach for artificial limbs, to develop multi-sensory and highly dynamic prototype exoskeletons and prostheses that enable its sensor-rich and data-intensive implementation, and to evaluate the resulting controller benefits in human subject experiments. The project combines two teams who will integrate their complementary expertise in legged dynamics and control, state estimation and learning, sensor fusion, mechatronic design, real-time control, and gait analysis to achieve these aims. The long-term goal of this research is to improve the quality of life for people who depend on artificial limbs for mobility and gait rehabilitation, with target populations ranging from lower limb amputees to stroke and spinal cord injured patients to older adults requiring mobility aids. The project directly relates to the NIBIB mission of improving health by development and acceleration of biomedical technologies. In particular, with its focus on next generation human-robot systems for improving gait assistance and rehabilitation, the project addresses NIBIB's strategic goal of developing innovative biomedical technologies that integrate engineering with the physical and life sciences to solve complex problems and improve health.

今天，即使是正常的运动也很难掌握依赖假体或 出行和康复的外骨骼，可穿戴机器人有清晰的了解 与今天相比，需要对人工四肢更具互动性控制的社区 克服这一限制。该项目具有这种交互式控制。一种网络物理方法 追求控制，其中人造肢体利用丰富的感官信息来连续 关于用户和环境的原因并以改进的方式适应了用户和环境 运动稳定性，鲁棒性和多功能性。拟议研究的具体目的是 为人工四肢建立这种网络物理方法，以发展多感官和高度动态 原型外骨骼和假体，可实现其传感器丰富和数据密集型实施，以及 评估人类主题实验中产生的控制器益处。该项目结合了两个 将在腿部动态和控制中融合其互补专业知识的团队，状态估计 以及学习，传感器融合，机电设计，实时控制和步态分析以实现这些目标 目标。 这项研究的长期目标是改善依赖人造的人的生活质量 四肢出行和步态康复，目标人群从下肢截肢到 中风和脊髓受伤的患者是需要活动能力的老年人。该项目与 尼比布通过生物医学技术的发展和加速来改善健康的使命。在 特别是，它重点放在下一代人类机器人系统上，以改善步态援助和 康复，该项目探讨了尼比卜开发创新生物医学的战略目标 将工程学与物理和生命科学整合以解决复杂问题的技术 改善健康。

项目成果

Artificial Neural Network-Based Activities Classification, Gait Phase Estimation, and Prediction

基于人工神经网络的活动分类、步态阶段估计和预测

• DOI: 10.1007/s10439-023-03151-y
• 发表时间: 2023
• 期刊: Annals of Biomedical Engineering
• 影响因子: 3.8
• 作者: Yu, Shuangyue;Yang, Jianfu;Huang, Tzu-Hao;Zhu, Junxi;Visco, Christopher J.;Hameed, Farah;Stein, Joel;Zhou, Xianlian;Su, Hao
• 通讯作者: Su, Hao