Design Recommendations for Tracking Speed and Direction of Linear Smooth Pursuit under Motion Conditions

Smooth pursuit represents an emerging form of eye-control human-computer interaction, offering excellent robustness and anti-interference capability. Nevertheless, it remains to be seen whether this technology can be used to replace traditional keyboard and mouse input in complex motion environments. In this study, we conducted a comprehensive evaluation of the usability of linear trajectory smooth pursuit eye-control interaction through a series of ergonomics experiments. These experiments were designed to create static environments and three distinct motion working conditions, namely vibration, shaking, and a complex condition that combined vibration and shaking. The Euclidean algorithm was employed as the tracking detection method to investigate the impact of varying tracking speeds (5°/s, 7.5°/s, 10°/s, 12.5°/s) and tracking directions (0°, 45°, 90°, 135°) on the performance of the simple choice task and the subjective load assessment of the interaction task in each motion environment. The results demonstrated that varying levels of tracking speed and direction maintained a satisfactory task completion rate for this interaction in motion conditions. The study proposed an optimal tracking speed of 12.5°/s and optimal tracking directions of 0° (horizontal) and 135° (upper left). These findings offer novel solutions for human-computer interaction in motion environments and serve as a crucial reference for the optimization of smooth pursuit eye-control interaction systems.