This paper introduces RobotCycle, a novel ongoing project that leverages Autonomous Vehicle (AV) research to investigate how cycling infrastructure influences cyclist behaviour and safety during real-world journeys. The project's requirements were defined in collaboration with key stakeholders (i.e. city planners, cyclists, and policymakers), informing the design of risk and safety metrics and the data collection criteria. We propose a data-driven approach relying on a novel, rich dataset of diverse traffic scenes captured through a custom-designed wearable sensing unit. We extract road-user trajectories and analyse deviations suggesting risk or potentially hazardous interactions in correlation with infrastructural elements in the environment. Driving profiles and trajectory patterns are associated with local road segments, driving conditions, and road-user interactions to predict traffic behaviour and identify critical scenarios. Moreover, leveraging advancements in AV research, the project extracts detailed 3D maps, traffic flow patterns, and trajectory models to provide an in-depth assessment and analysis of the behaviour of all traffic agents. This data can then inform the design of cyclist-friendly road infrastructure, improving road safety and cyclability, as it provides valuable insights for enhancing cyclist protection and promoting sustainable urban mobility.

翻译：本文介绍了一项名为RobotCycle的新型持续性项目，该项目利用自动驾驶汽车（AV）研究成果，探究真实骑行场景中自行车基础设施对骑行行为及安全性的影响。项目需求通过与关键利益相关方（城市规划者、骑行者及政策制定者）的协作共同确定，并据此设计了风险与安全评估指标及数据采集标准。我们提出一种数据驱动方法，基于通过定制可穿戴传感单元采集的多样化交通场景构建的新型丰富数据集。通过提取道路使用者轨迹，分析其中与环境基础设施要素相关的偏离模式，进而识别潜在风险或危险交互行为。驾驶特征与轨迹模式被关联至局部道路区段、驾驶条件及道路使用者交互行为，用于预测交通动态并标注关键场景。此外，项目借助自动驾驶研究的先进技术，提取高精度三维地图、交通流模式及轨迹模型，实现对全交通参与者的行为深度评估分析。该数据可为设计骑行友好型道路基础设施提供支撑，通过增强骑行者保护与促进可持续城市出行，切实提升道路安全性与骑行友好度。