AutHUMate

Adaptive User Training & HUMan-Technology Interaction StrAtegy for Enhanced Traffic SafEty

Future vehicle systems will only enhance safety if they are trusted, understood, and appropriately used by drivers. Yet, advanced driver assistance and automation technologies are often avoided, misunderstood, or misused due to limited system transparency or insufficient knowledge.

AutHUMate addresses these challenges by developing an adaptive, self-learning Human–Technology Interaction (HTI) strategy built on a closed sensing–action–learning loop. Therefore in-cabin and contextual data are combined into a driver-in-context model that interprets driver readiness relative to the environment. Based on this, the system adapts what, when, and how to communicate through multimodal interfaces. Each interaction feeds back into the loop, customising support, preventing mode confusion, and balancing trust to avoid both underuse and overreliance.

To foster appropriate use, AutHUMate pursues a dual training approach: (1) static, game-based learning to build baseline understanding, and (2) dynamic, in-car gamification that adapts in real time to underuse or overuse. User-centeredness is ensured through a Societal Readiness Assessment (SRA) framework with its modular mixed-methods toolkit that integrates quantitative metrics on i.e. trust and acceptance, qualitative feedback on e.g. fairness and clarity, and participatory formats such as co-creation workshops and expert dialogues. Validation is carried out in both high-fidelity simulators and real traffic environments, involving diverse user groups, such as young, elderly, and professional drivers