Supporting Driver Attention Toward Potential Hazards During Takeover: A Preliminary Result

Abstract

Conditionally automated vehicles still rely on drivers to retake control when something unexpected emerges. We cataloged hazard types from naturalistic driving data and used those insights to design a gaze-guidance takeover aid grounded in the N-SEEV attention model. In a driving simulator study, the highly salient guidance cut secondary-hazard collisions during takeovers, indicating that directing eyes toward relevant risks can improve transition safety.

Publication
In Proceedings of the Human Factors and Ergonomics Society Annual Meeting, 2024