SBIR Phase II: A Novel Human Machine Interface for Assistive Robots

这一小型企业创新研究(SBIR)第二阶段项目的更广泛影响/商业潜力旨在使全球目前失去肢体或残障的2亿多人受益。随着人口老龄化的快速增长和预期寿命的延长，迫切需要能够提高人们独立性和自给自足能力的辅助技术，使他们能够更长时间地呆在家里。拟议中的可穿戴传感器将使旨在帮助日常生活活动的机器人更加有效、负担得起和易于使用。除了使人们能够实现更高水平的功能和生活质量外，该传感器还可以促进对血流动力学模式所表现出的生理变化的基本了解，这可以用于更好地监控患者状态，并使临床医生和辅助设备制造商能够开发更个性化和更有针对性的解决方案。这个小型企业创新研究(SBIR)第二阶段项目旨在开发一种紧凑且低成本的光学传感器，用于检测残疾用户的手势命令，并将手势转换为辅助机器人。目前大多数辅助机器人采用的人机界面价格昂贵，而且固有的噪音，需要广泛的处理和用户培训。需要一种更实用、更直观、更可靠的解决方案，以更好地适应最终用户多样化且经常不断变化的情况。这项研究将致力于提高传感器作为可穿戴辅助机器人的可嵌入部件的可靠性、可用性和兼容性。传感器模块可以以菊花链的形式以不同的布置连接在一起，将被设计成最佳地监测手臂上的不同肌肉活动。先进的信号处理和机器学习技术将用于扩展现有的手势检测算法，并在日常设备使用中实现更稳健的性能，解决诸如同时检测多个命令和启用长期算法学习等实际问题。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase II project seeks to benefit more than 200 million people around the globe who are currently living with limb loss or impairment. With the rapid growth of an aging population and longer life expectancies, assistive technologies that can improve the independence and self-sufficiency of people, enabling them stay in their homes longer, are urgently needed. The proposed wearable sensor will be a step towards making robots designed to assist in activities of daily living more effective, affordable, and easy to use. In addition to empowering people to achieve higher levels of functionality and quality of life, this sensor may also further the fundamental understanding of physiological changes as manifested in hemodynamic patterns, which could be used to better monitor patient status and allow clinicians, as well as assistive device manufacturers, to develop more personalized and mindful solutions. This Small Business Innovation Research (SBIR) Phase II project aims to develop a compact and low-cost optical sensor for detecting gesture commands from disabled users and to translate the gestures to assistive robots. The human-machine interfaces currently adopted by most assistive robots are expensive and inherently noisy, requiring extensive processing and user training. A more practical, intuitive, and reliable solution is needed to better accommodate the diverse and often evolving conditions of end users. This research will focus on enhancing the reliability, usability, and compatibility of the sensor as an embeddable component for wearable assistive robots. Sensor modules that can be daisy-chained together in various arrangements will be designed to optimally monitor different muscle activities on the arm. Advanced signal processing and machine learning techniques will be used to expand the existing gesture detection algorithm and achieve more robust performance during daily device usage, addressing practical issues such as detecting multiple commands simultaneously and enabling long-term algorithm learning.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.