CAREER: Versatile Wearable Robots for Rehabilitation of Children with Gait Disabilities

这项教师早期职业发展（Career）资助旨在提高动力膝关节外骨骼的可用性和舒适性，旨在改善脑瘫儿童的步态生物力学和步行代谢经济。通过利用新型执行器和可穿戴传感器技术，PI将开发一种外骨骼系统，该系统可以随着孩子的成长而“成长”。PI还将设计一个任务识别系统（意图估计），以促进高级外骨骼控制。使用无模型自适应控制器增强了高级控制，该控制器可以提供用户特定的关节扭矩辅助。该项目将通过为一种安全、轻便、柔顺、智能并可在社区环境中使用的儿童步态康复新方法建立理论基础和实际实现，促进科学进步和国民健康。针对小学和高中学生的综合研究和教育计划以及创新的外展活动将促进具有全球竞争力的STEM劳动力，并促进妇女和代表性不足的少数民族参与STEM。进一步发展“软机器人动物园”的努力将提高公众的科学素养和对机器人的参与。该项目旨在使用一种新型的动力膝关节外骨骼来减少脑瘫儿童的步态障碍，这种外骨骼可以根据不断变化的任务条件和不断变化的用户意图来预测关节扭矩辅助的变化需求。研究了三个主要的研究目标：利用“准直接驱动”可穿戴机器人优化人机交互；基于可穿戴传感器技术的脑瘫儿童运动学活动分类与步态模式检测开发了一种基于强化学习（rl）的自适应辅助关节扭矩个性化控制器。综上所述，这些研究活动有望在现实世界中为患有身体残疾的儿童加强人机协作。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。

This Faculty Early Career Development (CAREER) grant seeks to enhance the usability and comfort of a powered knee exoskeleton that is designed to improve gait biomechanics and walking metabolic economy in children with cerebral palsy. By leveraging novel actuators and wearable sensor technology, the PI will develop an exoskeleton system that can "grow" in capability as the child grows. The PI will also design a task-recognition system (intent estimation) that facilitates high-level exoskeleton control. That high-level control is augmented using a model-free adaptive controller that can provide user-specific joint torque assistance. This project will promote the progress of science and advance the national health by establishing theoretical foundations for - and a physical realization of - a new approach to rehabilitating pediatric gait, one that is safe, lightweight, compliant, smart, and able to be used in community settings. An integrated research and education plan and innovative outreach activities targeted to elementary- and high-school students will promote a globally competitive STEM workforce and the participation of women and underrepresented minorities in STEM. Efforts to further develop a "Soft Robot Zoo" will increase public scientific literacy and engagement with robotics.This project seeks to reduce gait impairments in children with cerebral palsy using a novel powered knee exoskeleton that can anticipate changing needs for joint torque assistance based on changing task conditions as well as changing user intentions with regard to gait transitions. Three main research objectives are researched: an optimization of human-machine interaction using "quasi-direct drive actuated" wearable robotics; kinematic activity classification and gait mode detection in children with cerebral palsy using wearable sensor technology; and the development of a reinforcement learning (RL-based) adaptive controller for assistive joint torque personalization. Taken together, these research activities promise to enhance human/machine collaboration in a real-world setting for children contending with physical disability.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.