Home // AIVTS 2025, The First International Conference on AI-enabled Unmanned Autonomous Vehicles and Internet of Things for Critical Services // View article

An AI-Powered Aerial Navigation System for Dynamic Accessibility Aware Pathfinding

Authors:
Ayushi Dwivedi
Zahra Derakhshandeh

Keywords: Unmanned Aerial Vehicles; Drone-as-a-Service; Artificial Intelligence; Accessibility; Mobility Impairment.

Abstract:
Navigating indoor environments brings significant challenges for individuals with mobility impairments, such as wheelchair users, older adults, those with temporary injuries, and others requiring accessible pathways. A key barrier is the absence of reliable, real-time information about indoor layouts, accessibility features, and temporary obstacles. Existing navigation solutions often rely on static maps and outdated data, limiting their ability to address the dynamic and specific needs of users seeking accessible routes. To overcome these limitations, this study introduces an Artificial Intelligence (AI)-assisted drone-based navigation system that provides real-time guidance and adaptive support for individuals with mobility restrictions. We developed and integrated a custom object detection model into the aerial platform to identify accessibility features and environmental obstacles. In addition, a dynamic path-planning algorithm enables the drone to autonomously guide users through accessible routes, adjusting in real-time to environmental changes. The system reroutes users when unexpected obstructions arise, ensuring uninterrupted and reliable navigation to the targeted destination. We evaluated the system's performance through experiments in a controlled environment, demonstrating its effectiveness and potential for real-world applications.

Pages: 8 to 15

Copyright: Copyright (c) IARIA, 2025

Publication date: October 26, 2025

Published in: conference

ISBN: 978-1-68558-317-0

Location: Barcelona, Spain

Dates: from October 26, 2025 to October 30, 2025