Diederichs, FrederikFrederikDiederichsMarberger, ClausClausMarbergerHinder, VictorVictorHinder2022-03-112022-03-112010https://publica.fraunhofer.de/handle/publica/370658Advanced Driver Assistance Systems (ADAS) are expected to mitigate specific driver deficits and thus increase safety. For ADAS which do not take over any vehicle control the effect of the intervention is entirely dependent on the driver's reaction to the warning. The Human Machine Interface (HMI) concept reported in this paper is a result of the EU project SAFESPOT, funded within the 6th framework program and aiming to develop cooperative systems to increase traffic safety. The project goal is to develop a system that can issue warnings at different stages of criticality - from an early informative message to an imminent hazard warning shortly before a crash. Two experiments have already been carried out and published that describe elements of an innovative audio-visual HMI including new features like visual attractors distributed in the drivers field of view and acoustic hybricons (a fusion of earcons and auditory icons). This paper focuses on a third experiment that has been carried out in a driving simulator with 24 participants in December 2009 in order to evaluate the complete audio-visual SAFESPOT HMI. Comfort, Safety and Critical warnings where applied in three scenarios as Frontal Collision Warnings and Road Condition Status Warnings. The results show that Comfort warnings cause no sudden reactions like braking or lane change, however they lead to a speed reduction in uncertain environments like fog or slippery roads. Driving with warnings show clear advantages in terms of earlier speed reduction and fewer collisions for warned drivers. Subjective ratings are overall positive in terms of acceptance. Some participants perceived the warnings as being too obtrusive.enconference paper