University of Twente Student Theses
Sixthsense : outdoor collision avoidance assistant for the blind and visually impaired
Gougoulis Dimitriadis, E. (2022) Sixthsense : outdoor collision avoidance assistant for the blind and visually impaired.
Full text not available from this repository.
| Full Text Status: | Access to this publication is restricted |
| Abstract: | The present work aims to provide a novel outdoor collision assistant for the blind and visually impaired (BVI) population. Regardless of the rapid technological advancements of recent years, current BVI navigation and collision avoidance aids are still in a primitive stage. Our main goal is to create a system that can warn BVI users about nearby obstacles in a reliable and intuitive manner. To address this challenge, we investigate the integration of cutting-edge object detection and depth data in a small and mobile form factor. For this reason, a compact but powerful passive stereo depth camera (OAK-D) is used. In an attempt to limit the overwhelming quantity of information that previous navigation aid systems pass to the user, we explore the prioritization of collision alerts. More specifically, obstacle warnings are prioritized based on their estimated collision time with the BVI. Communication with the user is achieved via a smart vest, which allows him to continue using other widely-adopted navigation tools like the white cane. The smart vest comprises of a 3x3 grid of vibration actuators spread in its back area, indicating this way the location of approaching obstacles. Additionally, various obstacle proximity levels can be conveyed through the vibration intensity of the haptic feedback. For convenience, the camera hardware is mounted in the front area of the smart vest. The results of testing the developed system in a variety of situations suggest its ability to provide reliable outdoor collision assistance to the blind and visually impaired. |
| Item Type: | Essay (Master) |
| Faculty: | EEMCS: Electrical Engineering, Mathematics and Computer Science |
| Subject: | 54 computer science |
| Programme: | Computer Science MSc (60300) |
| Link to this item: | https://purl.utwente.nl/essays/93532 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page