The pathologic link between vision loss and physical inactivity has confounded doctors, scientists, engineers, and patients for centuries. Mobility losses are extremely variable and not only tied to significant compromises in quality of life, but also to an abysmal unemployment rate. One massive employment barrier related to mobility loss in urban areas is the ongoing struggle to utilize public transportation during daily commuting. While technological advances in data science and artificial intelligence continue to rapidly spawn new digital tools and electronic solutions, many of these new devices are incomplete answers that leave end users cognitively overwhelmed between device switching and task completion. This project will support foundational research needed to study low-vision behavior and develop more powerful wearables that can handle data-intensive processing, enabling parallel functionality. The project will afford VIS4ION, a revolutionary wearable platform that uses backpack-mounted sensors, advanced machine vision, wireless communications, and human-machine interfaces, the ability to perform ‘connected’ dynamic localization and navigation assistance for the visually impaired in complex urban environments. The project will lead to a healthier low vision population with more gainful employment, a framework for behavioral investigation in disability studies, guidelines for the design and delivery of navigation-focused wearables, students who are well-versed in multi-disciplinary and disability-focused research, and a blueprint for a smart and connected community that enhances economic vitality, safety, security, health and wellbeing, and overall quality of life.

This research will respond to the commuting challenges that stymie employment by creating new connections between visually impaired residents and their surrounding environment through innovations in science and engineering. The project envisions fundamental research in novel behavioral studies, optimized data streams, enhanced power management, and city-agency cooperation to fill a number of knowledge gaps, including (1) lack of a scientifically-principled, experimentally-based understanding of users’ needs and behavioral patterns, (2) limited ability to interpret real-world scenes through computer vision algorithms and excessive computational burden, (3) incomplete understanding of technical methods to balance local versus cloud analytics and manage power effectively, and (4) lack of hypothesis-driven studies with ecological validity that could scale-up to solve commuting challenges. A convergent research plan will fill these gaps and contribute key advancements in computer vision, machine learning, video compression, wireless transmission, human factors, and spatial positioning. By quantifying user needs regarding transportation and subsequently enhancing the safety profile of visually impaired travelers, gainful employment will improve, with secondary impacts on reducing health and economic burden.