TRUCE-AV
收藏资源简介:
TRUCE-AV数据集是一个用于自动驾驶中信任和舒适度估计的多模态数据集。该数据集收集了31名参与者在模拟器中进行完全自动驾驶时的实时信任投票和连续舒适度评分。同时,还记录了生理信号,如心率、注视模式和情绪,以及环境数据(例如,车辆速度、附近车辆的位置和速度)。此外,还进行了标准的前后驾驶问卷调查,以评估参与者对自动化的信任和整体福祉,从而能够将主观评估与实时响应相关联。该数据集使自适应AV系统的开发成为可能,这些系统能够根据用户的信任和舒适度水平进行动态响应,从而提高安全性、用户体验和以人为中心的车辆。
TRUCE-AV dataset is a multimodal dataset for trust and comfort estimation in autonomous driving. It collects real-time trust votes and continuous comfort ratings from 31 participants during fully autonomous driving sessions in a driving simulator. Meanwhile, physiological signals including heart rate, gaze patterns and emotional states, as well as environmental data (e.g., vehicle speed, positions and speeds of nearby vehicles) are recorded. Additionally, standard pre- and post-driving questionnaires are administered to evaluate participants' trust in automation and overall well-being, enabling the correlation between subjective assessments and real-time responses. This dataset enables the development of adaptive AV systems that can dynamically respond based on users' trust and comfort levels, thereby enhancing safety, user experience, and human-centric vehicles.




