Beyond infrastructure: exploring drivers of subjective cycling safety using directed acyclic graphs (DAG)

Engineering notes

Key topics: autonomous driving, perception, planning. See the paper for implementation details and experimental results.

Chinese explanation / 中文解读

中文解读待补充：本站会优先为端到端自动驾驶、BEV感知、3D目标检测、轨迹预测、路径规划、LiDAR感知等高价值论文补充中文说明。

Original abstract

Abstract A key reason why people refrain from cycling is a lack of perceived safety. Past research clearly demonstrates that safety perceptions are not only a result of different cycling infrastructures but reflect differences in individual and social factors. However, it remains unclear what causes these differences, as many studies on cycling safety do not pay sufficient attention to causal structures. This study deploys contrafactual causal analysis to screen existing evidence and develop a set of testable models. We use directed acyclic graphs (DAG) and statistically realize them via linear regression models to identify drivers of subjective cycling safety. In addition to gender and age, which are known to shape safety perceptions, we also investigate the impact of people’s social environment on cycling safety. Our analysis of telephone survey data from Germany (N = 1530) reveals that women feel less safe than men and safety perceptions vary across age groups, clearly supporting previous research. More importantly, the study shows that a pro-cycling social environment promotes subjective safety. Self-reported cycling competence acts as a central mediator across all three models while lower subjective safety among women—especially those with a migration background—reflects greater levels of anxiety. Implications for policy and planning such as infrastructure design for the most vulnerable and lifelong learning offers are also discussed.

Links and sources

• PDF from original source

Comments