Data underlying the master thesis: Predicting cycling risk at intersections with natural cycling data for speed-controlled e-bikes

This dataset contains data that was gathered during a master's thesis project. During the project, e-bike cycling data was gathered during a natural cycling experiment. Five women and five men aged 20 to 30 years were asked to participate in the experiment. Participants are asked to cycle according to how they would cycle in their daily life. The experiment consists of two phases. First, participants are instructed to cycle to a random destination in Delft. In the second phase, participants are instructed to cycle two predetermined routes, each time returning to the starting location of the experiment. This results in around 40-45 minutes of cycling data per participant. The sensor setup of the experiment is tailored toward collecting data that is similar to the data collection capabilities of modern IoT modules carried by e-bikes. Data was recorded by various sensors (a high-frequency GNSS antenna, IMU data and power pedals) and correlated against detailed open-access traffic accident data. In this dataset, you will find the raw sensor data, processed sensor data, and script used to process the cycling data.

本数据集源自某硕士学位论文项目的采集工作。项目期间，通过自然骑行实验采集电动自行车骑行数据。本次实验共招募10名参与者，年龄介于20至30岁之间，其中女性、男性各5名，要求其按照日常骑行习惯完成全部骑行任务。实验包含两个阶段：第一阶段，指示参与者骑行至代尔夫特（Delft）的任意随机目的地；第二阶段，指示参与者骑行两条预设路线，每次骑行结束后均返回实验起始地点。每位参与者的有效骑行数据时长约为40至45分钟。本实验的传感器配置专为采集与现代电动自行车搭载的物联网（Internet of Things, IoT）模块数据采集能力相匹配的数据而设计。实验数据由多种传感器采集，包括高频全球导航卫星系统（Global Navigation Satellite System, GNSS）天线、惯性测量单元（Inertial Measurement Unit, IMU）及功率脚踏板传感器，并与公开获取的详细交通事故数据集进行关联分析。本数据集包含原始传感器数据、经处理后的传感器数据，以及用于处理骑行数据的脚本。