Employing synesthesia-based warnings to enhance road safety during an automated driving

With the rapid employment of advanced driver assistance systems (ADAS), drivers will perform more varied and demanding tasks, which may result in reduced perception and response to emergencies. Existing studies have proved the effectiveness of multimodal warning signals and found that they can reduce drivers’ response time. However, most of these studies only focus on the superposition effect between different channels. There is a lack of research on the correlation between different channels and their impact on drivers’ workload level. Therefore, a multimodal warning signal model based on the theory of Audio-visual synesthesia is proposed in this paper, and the effects of this model on road safety in the scenario of automatic driving are explored through a controlled experiment of synesthesia and non-synesthesia signals. The results show that compared with the non-synesthetic signal, the multimodal warnings based on synesthesia can transmit the relevant information more quickly, and also reduce the workload level of the driver in the assisted driving scenario. The findings will aid the design of an early warning system for future autonomous vehicles.