Skip to content
HomeNewsresearch
News / RESEARCH

Preprint evaluates temporal safeguards for an eye-gaze wheelchair retrofit

A preprint describes computational testing of iMove while emphasizing that physical implementation and user evaluation remain outstanding.

Preprint evaluates temporal safeguards for an eye-gaze wheelchair retrofit

Rollivane editorial illustration; not a photograph of the reported event.

Based on the linked source; automatically prepared and checked against the original report.

System design

The iMove retrofit combines a commercial eye tracker, tablet interface, Arduino Uno, and two servomotors to actuate a wheelchair joystick. Preliminary testing produced eight successful selections in ten attempts, or 80.0%, with a 95% Wilson interval of 49.0% to 94.3%. The record identifies the study as a preprint, not peer-reviewed evidence.

Temporal supervision

The baseline controller activated on target entry and retained motion commands after target exit or interrupted communication. Researchers evaluated a supervisor using a 0.60-second motion dwell, 0.15-second input-freshness threshold, and 0.30-second receiver timeout. In deterministic fault scenarios, it neutralized commands 0.12 seconds after frozen input and 0.24 seconds after host or serial-link interruption, while the baseline remained active throughout the 3.5-second post-fault window.

Limits and stopping model

Ablation found that receiver timeout alone could not identify stale input when messages continued. At an assumed speed of 0.50 metres per second, a linear-deceleration model based on a preliminary three-to-four-second stopping interval produced a conditional stopping envelope of 0.78 to 1.13 metres. Physical implementation and evaluation with intended users remain necessary to establish operational performance.

Sources & further reading

Europe PMC — original report

Our editorial approach

CONTINUE EXPLORING

A little more useful reading.