SBIR Phase II: AI-Driven Orientation and Mobility System for the Blind

这项小企业创新研究第二阶段项目的更广泛影响/商业潜力是推进计算机视觉导航，以支持视力障碍人士的独立。在美国，大约有800万人患有失明或视力障碍，这严重降低了他们在新环境中独立导航的能力，从而限制了他们参与社会和经济生活。缺乏独立性直接导致在完成正规教育和加入劳动力方面处于不利地位。该项目旨在创造一个突破性的方向和移动系统，作为一个应用程序，叙述环境，并提供在室内环境中记录和追溯路线的能力。核心技术创新是一种多模态神经网络架构，能够学习鲁棒空间表示并有效地在移动设备上运行。拟议中的项目将引入一种新的人工智能（AI）驱动的定向和移动系统，以帮助视力受损的个人完成关键的导航任务。该方案将从智能手机摄像头中生成与上下文相关的、面向任务的空间信息。研究任务包括：(1)开发优化和数据增强技术，以提高各种移动设备的运行时性能；(2)开发环境的口头叙述以及路线创建和跟踪；(3)与潜在用户共同设计和测试。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。

The broader impact/commercial potential of this Small Business Innovation Research Phase II project is to advance computer vision for navigation to support independence for individuals with visual impairment. Roughly eight 8 million individuals in the US live with blindness or visual impairment, which severely reduces a person’s ability to independently navigate in new spaces and consequently limits participation in social and economic life. The lack of independence directly leads to disadvantages in completing formal education and joining the workforce. This project is aimed at creating a breakthrough orientation and mobility system as an application narrating environments and providing the ability to record and retrace routes in indoor environments. The core technological innovation is a multimodal neural network architecture capable of learning robust spatial representations and efficiently operating on mobile devices. The proposed project will introduce a novel artificial intelligence (AI)-driven Orientation and Mobility System to assist individuals with visual impairment in crucial navigation tasks. This proposal will generate contextually relevant, task-oriented spatial information from smartphone cameras. The research tasks include: (1) Development of optimization and data augmentation techniques for improved runtime performance for a broad range of mobile devices; (2) Development of verbal narration of environments as well as route creation and tracking, (3) Co-design and testing with potential users.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.