Eye-Tracking Calibration to Control a Cobot

Abstract

The present study pursues to determine the optimal operation range of a specific screenbased eye-tracker, the Tobii X2-30, regarding the variation of precision and accuracy in measures. Furthermore, a connection setup to operate a collaborative robot (cobot) Omron TM5-700 by means of this eye-tracker will be presented. The possibility to operate a collaborative robot by gaze can be used as a third arm, which allows human beings to do more sophisticated activities, as well as making the manipulation of dangerous or perilous substance easier and safer. When developing new technological tools, we have mainly two options. The first one consists on a specifically designed hardware. While in this option, the engineer has full control over the device and can fit it to the specific requirements; in general, it will be a time consuming and expensive development. A second drawback is the limited possibility of researchers from other countries to construct an exactly equal device and replicate the experiments. The second option consists on adopting an existing commercial hardware, which probably has not been designed for the specific application in mind. The main advantage is the easy adaptation of this solution by other researchers, who only need to purchase the same commercial device and follow the recommendations. However, the main drawback of this approach is that the developers must test the device and check that it can be used for the new application. The goal of this paper is to test a commercial device and provide usability recommendations for a new application such is the movement of a robotic arm using eye-tracking. This paper includes the results from three experiments, which assess the final conclusion on the best performance positioning of the user regarding the Tobii X2-30 eye-tracker, in x, y, and z coordinates. When it comes to its implementation with the cobot, the outcome of a practical demo and experimental setup is also presented. This last one consists of accuracy measurements, where the control of the position of the cobot is defined by means of gaze, which defines a set of points in (x,y) plane. Later on, the robot picks up an ink-pen and draws a graph in a piece of paper. This drawing involves connecting these pre-defined dots by straights lines. To this end, a set of figures (parallelogram, pentagon, etc.) have been acquired and compared with the desired printed images on the PC screen.

Evaluating computer science students reading comprehension of educational multimedia-enhanced text using scalable eye-tracking methodology

“…The library aims to address the limitations of traditional eye-tracking devices by offering a cost-effective and scalable solution for assessing gaze behaviors in educational settings, particularly for neurodivergent students (Asghari et al, 2023). However, webcam-based eye tracking, in general, faces challenges related to spatial accuracy, calibration validity, and gaze prediction methods (Faura-Pujol et al, 2023). Efforts have been made to improve data quality by estimating spatial offset and enhancing the correlation with Areas of Interests (AOIs) defined over stimuli (Abhaya et al, 2022).…”