IoT Empowered Assistive Device for Partially Sighted People
Madhumathi, K. and Akila, A. (2026) IoT Empowered Assistive Device for Partially Sighted People. In: 2026 International Conference on AI-Driven Smart Systems and Ubiquitous Computing (ICAUC), Pathum Thani, Thailand.
14th_IEEE_International_Symposium_on_Applications_of_Ferroelectrics__2004__ISAF_04__20041778404933855 (1).pdf
Download (1MB)
Abstract
The partially sighted or Visually Impaired (VI) are greatly disadvantaged when it comes to safe navigation and a self-sufficient implementation of daily tasks both indoors and outdoors as they have limited awareness of the environmental obstacles around them. The given paper provides an IoT-enabled assistive device that will facilitate autonomous and safe navigation. The proposed system is a combination of systems that incorporate hardware and software wherein an array of ultrasonic sensors is used to acquire real-time distance information in various directions. As measurements by sensors are prone to acoustic and electrical, and environmental noises, a Kalman filter is used to reduce noise and increase the accuracy of the data. Fusion of the filtered sensor data with a hierarchical fuzzy logic inference system is then done to estimate proximity to obstacles. Besides it, a lightweight adaptive prediction algorithm is also presented that predicts the obstacle positions with the user navigation route depending on the time dependence between sensor data and proactively guides the user. The wireless communication with the help of IoT enables a smooth exchange of data between a sensing, processing, and feedback module and provides the user with real-time audio or haptic notifications.
| Item Type: | Conference or Workshop Item (Paper) |
|---|---|
| Subjects: | Computer Science Engineering > Computer Vision |
| Domains: | Computer Science |
| Depositing User: | Mr IR Admin |
| Date Deposited: | 10 May 2026 15:18 |
| Last Modified: | 10 May 2026 15:23 |
| URI: | https://ir.vistas.ac.in/id/eprint/15240 |
Dimensions
Dimensions