Trolley Remote Control System Using Eye Motion Tracking

In this paper, a trolley remote control system based on eye tracking technology is proposed that controls the direction of the trolley's movement according to the gaze direction of the human eye. The system's graphical user interface contains eight arrows pointing in different directions that function as the gaze targets for moving the trolley in the corresponding direction. This paper compares two algorithms for detecting fixation points. One method clusters the gaze position of the human eye recorded by the eye tracker in the preset time window, and the other method directly uses the fixation point determined by the eye tracker's own software to construct a fixation heat map and then perform image processing. Through experiments, it was found that the average target detection accuracy of the system using the clustering algorithm (99.78%) is 3.13% higher than that of the system using the second method and the difference between the performances of the algorithms was found to be significant. The results demonstrate that in the trolley remote control scenario, the use of eye tracking technology with a clustering algorithm can realize efficient and natural human-computer interaction. Moreover, such a system has good sensitivity and stability.