GazeRace: Revolutionizing Remote Piloting with Eye-Gaze Control

TL;DR

Problem: remote piloting often imposes cognitive and physical burden on operators. Approach: GazeRace combines real-time iris/eyelid tracking with the MediaPipe CV pipeline to convert gaze into nine discrete drone-control actions mapped to pitch, roll, yaw, and throttle within a ROS-Gazebo SITL framework. Findings: the system achieved an 18% reduction in flight-path length with competitive speeds, and subjective assessments show lower workload and higher hedonic quality for gaze control compared to a traditional remote controller. Significance: this hands-free, eye-gaze interface offers a practical path to more intuitive, accessible, and robust remote piloting with potential extensions to real drones and AR-enhanced navigation.

Abstract

This paper presents GazeRace, a novel system that leverages eye-tracking technology for intuitive drone control. Using the MediaPipe library, the system translates eye movements into precise drone commands, enabling effective remote piloting. In testing, GazeRace demonstrated an 18% reduction in drone trajectory length while maintaining competitive speed with traditional controls. The results suggest that this approach enhances control accuracy and reduces user frustration, offering a significant advancement in the field of human-computer interaction and drone navigation.

Figures (7)

• Figure 1: GazeRace interface for guiding racing drone: (a) MediaPipe interface with predefined right eye landmarks. (b) Experimental setup for the evaluation of eye control interface. (c) Drone race track in a Gazebo simulation environment.
• Figure 2: The pipeline of the drone mixed control approach based on the remote controller and the developed eye-tracking interface.
• Figure 3: Face mesh reconstruction with MediaPipe framework utilized for eye-tracking based drone control of GazeRace.
• Figure 4: Ten actions ( up, down, left, right, far-left, far-right, wide, squint, center, and one eyebrow action) utilized for a low-delay drone motion control.
• Figure 5: Set of recorded trajectories by the remote controller (red lines), and by eye-tracking interface (blue lines) with shortest path and flight time.