A smart cane-based indoor positioning system for visually impaired people
File(s) Wewalk_IEEE_Sensors (1).pdf (6 MB)
Accepted version
Author(s)
Type
Journal Article
Abstract
Autonomous navigation is a crucial aspect of the visually impaired community’s pursuit of independence and social equity. To achieve this goal, Visually Impaired People (VIP) require a navigation and wayfinding system that works both outdoors and indoors. While outdoor navigation can rely on ubiquitous signals such as Global Navigation Satellite System (GNSS), indoor navigation presents a significant challenge, especially in unfamiliar environments. Despite over two decades of research and development, the issue of delivering a high performance and cost-effective indoor positioning solution for VIPs remains unresolved. Bluetooth-based Angle of Arrival (AOA) positioning is a cost-effective method that offers accuracy ranging from upper decimetres to meters. However, it still faces challenges in detecting and mitigating outliers caused by non-line-of-sight (NLOS) and multipath effects during signal propagation. In light of these challenges, this paper proposes a novel dead reckoning aided AOA Bluetooth indoor positioning system that utilises a smart cane. The developed model takes into account the swinging characteristic of the cane, which is integral to visually impaired individuals’ mobility. Furthermore, the paper incorporates a novel outlier detection and exclusion layer to reduce the impact of NLOS and multipath effects. The proposed system’s performance was evaluated through a series of experiments, and the positioning error distribution was analysed. The results showed that the accumulated probability of errors less than 0.5 m was 93.56%, and 99.64% of errors were within 0.7 m. With the demonstrated high level of accuracy, the proposed system presents a promising solution for indoor navigation for VIPs.
Date Issued
2025-04-01
Date Acceptance
2025-02-01
Citation
IEEE Sensors Journal, 2025, 25 (7), pp.12085-12096
ISSN
1530-437X
Publisher
Institute of Electrical and Electronics Engineers
Start Page
12085
End Page
12096
Journal / Book Title
IEEE Sensors Journal
Volume
25
Issue
7
Copyright Statement
Copyright © IEEE 2025. This is the author’s accepted manuscript made available under a CC-BY licence in accordance with Imperial’s Research Publications Open Access policy (www.imperial.ac.uk/oa-policy)
Publication Status
Published
Date Publish Online
2025-02-17
