Designing AR-HUD Warnings for Safer Automated Vehicle Takeovers Among Older Drivers

Age-related declines in perception and attentional shifting may increase the difficulty of responding to takeover requests (TORs) in conditionally automated vehicles, particularly for older drivers. This study examined whether augmented reality head-up display (AR-HUD) warnings designed with different semantic mapping levels were associated with differences in takeover performance and subjective experience. Thirty younger and thirty older drivers completed emergency takeover scenarios under high and low semantic mapping AR-HUD warnings, and voice-only warning conditions. Dependent measures included situation awareness, subjective workload, system usability, and takeover reaction time. The results showed that younger drivers demonstrated higher situation awareness than older drivers. Compared with the low semantic mapping design, the high semantic mapping was associated with higher situation awareness, lower interference-related workload, and better usability. No significant differences were found in takeover reaction time. The study provides preliminary support for semantically and spatially interpretable AR-HUD warnings configurations in age-inclusive takeover interface design.