Safe and Efficient E-wayfinding (SeeWay) Assistive Navigation for the Visually Impaired
-
2022-10-10
-
Details:
-
Creators:
-
Corporate Creators:
-
Corporate Contributors:
-
Subject/TRT Terms:
-
Resource Type:
-
Geographical Coverage:
-
Edition:Final Report (August 2019 - July 2022)
-
Corporate Publisher:
-
Abstract:Despite its challenges, independent travel for blind and visually impaired (BVI) individuals is an essential component of quality of life, enabling travel to work and recreational activities. Autonomous vehicle technologies have the potential of meeting these challenges. However, efficiently and safely guiding BVI travelers between indoor environments and vehicles outdoors remains a key obstacle. In the future transportation system, assistive navigation technologies, connecting BVI travelers and vehicles, will be of extraordinary importance for BVI individuals in the context of social justice and health care/public health. Conventional research is mainly based on robotic navigation approaches through localization, mapping, and path-planning frameworks. They require heavy manual annotation of semantic information in maps and its alignment with sensor mapping. Inspired by the fact that we human beings naturally rely on language instruction inquiry and visual scene understanding to navigate in an unfamiliar environment, this study proposes a novel vision-language model-based approach for BVI navigation. It does not need heavy-labeled indoor maps and provides a Safe and Efficient E-Wayfinding (SeeWay) assistive solution for BVI individuals. The system consists of a scene-graph map construction module, a navigation path generation module for global path inference by vision-language navigation (VLN), and a navigation with obstacle avoidance module for real-time local navigation. The SeeWay system was deployed on portable iPhone devices with cloud computing assistance for the VLN model inference. The field tests show the effectiveness of the VLN global path finding and local path re-planning. Experiments and quantitative results reveal that heuristic-style instruction outperforms direction/detailed-style instructions for VLN success rate (SR), and the SR decreases as the navigation length increases.
-
Format:
-
Funding:
-
Collection(s):
-
Main Document Checksum:
-
Download URL:
-
File Type: