Eye-Movement AI Detection for Driver Notification Filtering

The system applies different notification filtering rules to different users based on their specific situation. The processing circuit determines whether each user is actively driving or not, and applies notification filtering only to actively driving users, while allowing notifications to reach passengers and autonomous vehicle drivers. This localized differentiation resolves the contradiction by making the notification system adaptive to individual user contexts rather than applying a blanket filtering rule to all users in moving vehicles.

The system performs a preliminary learning phase during which the processing circuit collects eye movement data from a specific user over time and determines that user's individual threshold. This preliminary action establishes a personalized baseline before actual driving state detection begins, ensuring that the threshold reflects each user's natural eye movement characteristics. This resolves the contradiction by preventing false positives for users with naturally high gaze frequency while maintaining high detection sensitivity through individualized calibration.

The system automatically performs the calibration process without requiring manual intervention or complex user setup. The processing circuit autonomously collects eye movement data during a learning phase, analyzes the patterns, and determines individualized thresholds for each user. This self-service approach resolves the contradiction by achieving high detection accuracy through individualized calibration while minimizing system complexity and user burden, as the calibration happens automatically in the background during normal vehicle operation.