人类-机器人信任数据集
收藏资源简介:
本研究构建了一个名为'人类-机器人信任数据集'的专用数据集,由美国弗吉尼亚大学工程与应用科学学院的研究人员创建。该数据集通过一个旨在引发信任和不信任实例的人类-机器人监督互动研究得来,包含了30名参与者的生理测量(如皮肤电活动、血容量脉冲、皮肤温度)、注视位置以及面部表情动作单元强度数据。这些数据与参与者对机器人伙伴的信任度自我报告相结合,用于训练机器学习模型,以识别对机器人信任的客观数据指标。数据集的应用领域是人类-机器人交互,旨在解决实时监测和预测人类对机器人伙伴信任度的问题。
This study constructs a specialized dataset named 'Human-Robot Trust Dataset', created by researchers from the Engineering and Applied Science School at the University of Virginia. The dataset is derived from a human-robot supervised interaction study designed to elicit instances of trust and distrust, and includes physiological measurements (such as skin conductance activity, blood volume pulse, skin temperature), gaze positions, and intensity data of facial expression action units from 30 participants. These data are combined with the participants' self-reported trust in the robotic partner to train machine learning models that can identify objective data metrics of trust in robots. The application domain of the dataset is human-robot interaction, aiming to address the issue of real-time monitoring and prediction of the trust level of human partners in robots.




