Advancements in Smart Navigation Systems: A Review for Visually Impaired and Elderly Individuals

¹ Imperial College of London, Department of Computing, London, United Kingdom
^2,3 Heriot-Watt University, School of Engineering and Physical Sciences, Dubai, United Arab Emirates
⁴ Karpagam College of Engineering, Coimbatore, India

^* Corresponding author: atk2001@hw.ac.uk

Published online: 19 December 2025

Abstract

Navigation assistance is becoming increasingly important for visually impaired and the elderly, as traditional tools often lack sufficient feedback for safe, independent mobility. With recent advancements in assistive technology, particularly in smart canes, the internet of things (IoT) is integrated with artificial intelligence (AI) to enable real-time environmental awareness. This paper traces the evolution of simple white canes into advanced devices fitted with sensors and computer vision technology, guiding users through real-time audio and sensory feedback. Additionally, a comparative analysis of object detection models-Faster Regional Convolutional Neural Network (Faster R-CNN), You Only Look Once (YOLO), and Single Shot MultiBox Detector (SSD)-used in assistive devices is presented to aid visually impaired and elderly users in detecting hazards. The accuracy, processing time, and real-time feasibility of each model were evaluated. Simulation results indicated that YOLO-v8 outperformed Faster-RCNN and SSD, achieving the highest accuracy of 92.6% under the same testing conditions. In summary, the integration of IoT and AI has significantly enhanced the functionality of navigation assistance devices, improving safety, independence, and quality of life for the elderly and people with vision impairment.