Supplementary data for the article: Personal Preference or Intrinsic Quality? A Study on Bicycle Handling Quality

This dataset contains post-processed data from an experimental study investigating whether bicycle handling quality is a matter of personal preference or an intrinsic, measurable property of the vehicle. Nineteen cyclists performed a controlled riding task on an instrumented cargo bicycle with four adjustable configurations (A, B, C, D), modified by changing the position of a 3 kg mass on the steering assembly. This alteration affected the steering inertia and handling dynamics while remaining visually concealed from participants.<br>The experiment took place indoors on a flat surface using a predefined track including straight-line riding, a U-turn, and a slalom. Each participant completed 8 trials, experiencing all pairwise transitions between configurations in a balanced order. They were instructed to maintain a constant speed of 5 km/h and to ride as straight as possible, minimizing deviations of the front wheel contact point.<br>Performance was objectively measured using three inertial measurement units (IMUs) embedded in smartphones (Nokia 3.4), sampling at 200 Hz.The performance metric, normalized front wheel path length, was computed using ground plane kinematics and bicycle's parameters.<br>Workload was assessed using a relative version of the NASA-TLX questionnaire, focusing on five dimensions: Mental Demand, Physical Demand, Temporal Demand, Effort, and Frustration. After every two consecutive trials, participants reported perceived changes and estimated percentage differences, later normalized per participant to reduce inter-subject variability.<br>Participants were 19 adults (4 female, 15 male) with varying self-assessed cycling expertise (Medium: 6, High: 5, Very High: 7). All provided informed consent. The study was approved by the Human Research Ethics Committee of TU Delft.<br>The dataset includes performance metrics, workload assessments, trial sequences, and configuration parameters.

本数据集源自一项实验研究的后处理数据，该研究旨在探究自行车操控品质究竟属于个人偏好范畴，还是车辆本身固有的、可量化的属性。19名骑行者在一台配备传感测量装置的货运自行车上完成受控骑行任务，该自行车设有四档可调配置（A、B、C、D），通过调整转向总成上3kg配重的位置实现配置调节。此改动会改变转向惯量与操控动力学特性，但不会让参与者察觉视觉变化。
实验在室内平坦场地开展，使用包含直线骑行、U型转弯与绕桩的预设赛道。每名参与者完成8次骑行，以平衡顺序体验所有配置间的两两切换。实验要求参与者保持5km/h的恒定车速，尽可能保持直行，最大限度减小前轮接触点的偏移量。
性能通过三部嵌入智能手机（Nokia 3.4）的惯性测量单元（IMU）进行客观测量，采样频率为200Hz。性能指标为归一化前轮路径长度，通过地面平面运动学数据与自行车参数计算得到。
工作负荷采用相对版NASA-TLX问卷进行评估，涵盖心理需求、体力需求、时间需求、努力程度与挫败感五个维度。每完成连续两次骑行后，参与者需报告感知到的配置变化并估算百分比差异，后续会针对每名参与者进行归一化处理，以降低个体间差异。
本次实验共有19名成年参与者（4名女性，15名男性），其自评骑行专业水平各不相同：中等6人、高等5人、极高7人。所有参与者均签署了知情同意书，本研究已获代尔夫特理工大学（TU Delft）人类研究伦理委员会批准。
本数据集包含性能指标、工作负荷评估结果、骑行序列数据与配置参数。