Outdoor Localization Using BLE RSSI and Accessible Pedestrian Signals for the Visually Impaired at Intersections.

Area of Science:

• Human-Computer Interaction
• Assistive Technology
• Signal Processing

Background:

• Navigating intersections is a significant challenge for blind and visually impaired (BVI) individuals.
• Current audible signals at pedestrian crossings can cause confusion, especially in complex urban environments.
• Existing location tracking methods struggle with outdoor noise and accuracy limitations.

Purpose of the Study:

• To propose and evaluate a novel assistive system, the Crossing Assistance System (CAS).
• To leverage Bluetooth Low Energy (BLE) Received Signal Strength Indicator (RSSI) for accurate indoor and outdoor localization.
• To overcome the limitations of traditional methods for BVI pedestrian navigation.

Main Methods:

• Developed CAS utilizing BLE RSSI for user location tracking.
• Conducted two studies at a real-world intersection to demonstrate outdoor RSSI tracking feasibility.
• Employed k-nearest neighbors (kNN) and Support Vector Machine (SVM) algorithms for data analysis and classification.

Main Results:

• Achieved 99.8% accuracy in localizing users within four zones using kNN in the first study.
• The SVM method, using various RSSI features over a 5-second window, achieved 97.7% accuracy in localizing users within nine zones.
• Demonstrated the effectiveness of RSSI-based tracking in overcoming outdoor noise interference.

Conclusions:

• The CAS system shows high accuracy in localizing BVI users in complex environments.
• Findings have significant implications for the design of wearable technology and traffic infrastructure for blind pedestrians.
• The proposed system offers a promising solution for enhancing the independence and safety of BVI individuals during navigation.