Towards Perception-based Collision Avoidance for UAVs when Guiding the Visually Impaired

2025-06-17

AI Summary

To address the challenge of UAV-assisted obstacle avoidance for outdoor navigation by visually impaired persons (VIPs), this paper proposes a real-time避障 system integrating global path planning with vision-based perception. Methodologically, we introduce a novel geometric dual-agent (UAV + VIP) collision-avoidance model tailored to human-UAV co-navigation scenarios, and design a multi-DNN collaborative framework that jointly enforces bidirectional safety constraints under dynamic environments. The system unifies GPS- and map-based global trajectory planning, onboard visual perception, and geometric local path replanning. Evaluated in realistic university campus settings—including sidewalks, parked vehicles, and crowded streets—the system achieves real-time, bidirectional collision avoidance, significantly enhancing navigation safety and robustness. Our core contributions are: (1) a geometric dual-agent modeling paradigm for human-UAV interaction, and (2) a multi-DNN-driven bidirectional safety coordination mechanism.

Technology Category

• Humans and AI: Human-Aware Planning and Behavior Prediction
• Intelligent Robots: Motion and Path Planning
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Autonomous navigation by drones using onboard sensors combined with machine learning and computer vision algorithms is impacting a number of domains, including agriculture, logistics, and disaster management. In this paper, we examine the use of drones for assisting visually impaired people (VIPs) in navigating through outdoor urban environments. Specifically, we present a perception-based path planning system for local planning around the neighborhood of the VIP, integrated with a global planner based on GPS and maps for coarse planning. We represent the problem using a geometric formulation and propose a multi DNN based framework for obstacle avoidance of the UAV as well as the VIP. Our evaluations conducted on a drone human system in a university campus environment verifies the feasibility of our algorithms in three scenarios; when the VIP walks on a footpath, near parked vehicles, and in a crowded street.

Problem

• Develop perception-based collision avoidance for UAVs aiding visually impaired
• Integrate local and global path planning for outdoor urban navigation
• Evaluate multi-DNN obstacle avoidance in real-world pedestrian scenarios

Innovation

• Perception-based path planning for local navigation
• Multi DNN framework for obstacle avoidance
• Integration of GPS and maps for global planning