Evaluation of traffic signs and markings considering human drivers and automated vehicles

This is the authors' abstract. We don't add key points for this paper.

Traffic signs and markings (TSMs) are necessary components of roadway infrastructure. Currently, the design of TSMs primarily caters to the needs of human drivers, without considering the requirements of automated vehicles (AVs). This study proposes a readiness evaluation framework for TSMs that assesses the effectiveness of their physical properties and roles in the driving tasks from the perspectives of both human drivers and AVs and compares the perspectives based on human drivers and AVs perceptual differences. This study chose a crash-prone roadway as the research object. For human drivers, subjective assessment is employed to evaluate the physical attributes and driving task effects of TSMs. For AVs, video recognition and simulation methods are used to assess the physical attributes and driving task effects of TSMs, respectively. The results indicate that there are differences in the perception of TSMs between human drivers and AVs. This finding emphasizes the need for future transportation infrastructure design to consider and understand the distinct requirements of autonomous vehicles and human drivers. Optimizing transportation infrastructure should ensure that they can be accurately recognized and understood by AVs. This research provides direction and guidance for the future design and optimization of TSMs that meet the safety needs of both human drivers and automated vehicles driving together.