Інтелектуальне управління режимами праці водіїв в умовах критичної нестабільності

Abstract

This study addresses AI-driven optimization of driver labor regimes amidst critical instability. Standard AETR compliance is hindered by wartime infrastructure damage and curfews. We propose a system integrating Computer Vision (PERCLOS) with CAN-bus data fusion for cross-modal fatigue validation. Utilizing Edge Computing ensures operational resilience during power outages and communication blackouts. Predictive analytics determine Safe Hubs for rest, accounting for external risks and strict timing. This approach transforms passive monitoring into an adaptive assistance system, significantly enhancing logistics stability and driver safety in high-risk environments.