NSF Convergence Accelerator: Track H: Mobility Independence through Accelerated Wheelchair Intelligence

今天有数百万潜在的电动轮椅司机，他们很难独立行动，甚至完全不可能。其结果是剥夺了残疾人及其照顾者在社区和劳动力中的独立性和参与度。本项目旨在从根本上改变从人体获取控制输入并将其传递给机器的方式，并利用实用的机器智能来提高轮椅的安全性并促进轮椅的独立操作，从而加速电动轮椅的可及性和实用性。轮椅行业尚未充分利用人工智能和驾驶辅助工具来帮助确保安全性和有效性。到2020年，随着第一个商业化的、fda注册的电动轮椅驾驶员辅助系统（LUCI）进入市场，这种情况开始发生变化。然而，他们的（仅减法）辅助与机器人公司甚至汽车驾驶员辅助实现的自动化之间仍然存在差距。从主动绕开障碍物转向到半自动导航，主动驾驶辅助系统的潜在影响是巨大的。该项目将向市场推出首个电动轮椅主动驾驶辅助系统。它还将继续开发和倡导数字和开放轮椅通信标准，并开发符合个人能力和偏好的灵活控制界面。有了这些创新，该项目将极大地扩大严重运动障碍患者获得独立行动的机会——这对提高生活质量和参与劳动力、对残疾人及其照顾者都具有重要意义。这个融合加速器的合作伙伴包括提供实际操作和指导原型设计的临床专业知识的利益相关者，以及实现和稍后翻译该设计的工程师。合作伙伴包括非营利组织（AbleGamers、Shirley Ryan AbilityLab）、学术界（西北大学）和工业界（LUCI Mobility、Function Engineering、LifeDrive）。以下研究重点将完成这项工作：-重点1：通过轮椅智能提高安全性和独立性，提供主动和完全自主的驾驶辅助。我们将在商用传感器轮椅安全系统（LUCI）内，向市场推出主动驾驶辅助系统。-推力2：通过倡导采用数字和开源API（轮椅数字接口）作为轮椅控制系统和控制输入之间通信的ISO/RESNA标准，使轮椅输入机制现代化和民主化，为使用任何数字接口来驱动电动轮椅打开大门。-目标3：通过开发一种灵活的多模式控制界面，弥合可以操作操纵杆和不能操作轮椅的司机之间在轮椅控制方面的差距，这种控制界面符合用户的要求，可以从他们身体的几乎任何地方捕获控制信号，并且可以与现有的轮椅和计算机控制界面协同使用。辅助轮椅市场已经准备好整合这些先进的功能，特别是患有退行性疾病的人，迫切需要这种大大扩大的辅助范围。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。

There are millions of potential power wheelchair drivers today for whom independent mobility is difficult, or even entirely impossible. The result is a deprivation-for persons with disabilities, and their caregivers-of independence and engagement within the community and workforce. This project proposes to accelerate the accessibility and utility of power wheelchairs by radically changing how control inputs are captured from the human body and transferred to the machine, and by leveraging practical machine intelligence to enhance safety and facilitate independent wheelchair operation. The wheelchair industry has yet to fully harness AI and driver assistance tools to help ensure safety and efficacy. In 2020 this began to change, as the first commercially-available, FDA-registered driver assistance system for power wheelchairs entered the market (LUCI). There remains a gap however between their (subtractive-only) assistance and the automation achieved by robot companies or even automobile driver assistance. From active steering around obstacles to semi-autonomous navigation, the potential impact for active driving assistance is tremendous. This project will bring to market the first active driving assistance system for power wheelchairs. It also will continue to develop and advocate for a digital and open wheelchair communication standard, and develop a flexible control interface that conforms to an individual's abilities and preferences. With these innovations, the project will greatly expand access to independent mobility by those with severe motor impairments-which is significant for both for quality of life and participation in the workforce, and by both persons with disabilities and their caregivers. The partners on this convergence accelerator include stakeholders to provide hands-on use and clinical expertise that guides the prototype design, and engineers to realize and later translate that design. The partnership spans non-profits (AbleGamers, Shirley Ryan AbilityLab), academia (Northwestern University), and industry (LUCI Mobility, Function Engineering, LifeDrive).The following research thrusts will accomplish this work:- Thrust 1: Advancing safety and independence with wheelchair intelligence that provides active, as well as fully autonomous, driving assistance. We will bring to market active driving assistance within a commercial sensor-based wheelchair safety system (LUCI).- Thrust 2: Modernizing and democratizing the wheelchair input mechanism, by advocating for the adoption of a digital and open-source API (the Wheelchair Digital Interface) as the ISO/RESNA standard for communication between wheelchair control systems and control inputs-opening the door to any digital interface being used to drive a power wheelchair. - Thrust 3: Bridging the gap in accessible wheelchair control between drivers who can operate a joystick and those who cannot, through the development of a flexible and multimodal control interface that conforms to the user and captures control signals from nearly anywhere on their body, and can be used in concert with existing interfaces for both wheelchair and computer control.The assistive wheelchair market is ready for the integration of these advanced capabilities, and persons with degenerative diseases in particular need this greatly expanded scope of assistance without delay.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.