From Action Perception to Joint Actions: Learning from Joint Handover Actions of Human Dyads for Robotic Actions and Human-Robot interactions (A01)

从动作感知到联合动作：从人类二元组的联合切换动作中学习机器人动作和人机交互（A01）

• 批准号: 437121936
• 金额: --
• 依托单位: Institut für Sportwissenschaft
• 依托单位国家: 德国
• 项目类别: Collaborative Research Centres
• 财政年份: 2020
• 资助国家: 德国
• 起止时间: 2019-12-31 至 2022-12-31
• 项目状态: 已结题
• 来源: https://gepris.dfg.de/gepris/projekt/437121936?language=en
• 关键词: Action; Perception; Joint; Actions; Learning

The objective of this project is to investigate joint action coordination in human dyads during handover tasks. Insights into joint action control can advance control mechanisms for joint actions in mixed (i.e., robot-human) dyads. In our experiments, we measure kinematics of reach and grasp as well as grasp force dynamics during handover tasks when systematically varying the composition of the dyads (e.g., age of the partners) and properties of the objects (WP1-3). Advanced analysis (e.g., dimensionality reduction) will be used to understand reach and grasp kinematics (WP4-6) and, therefore, predict action consequences (e.g., handover position, force scaling), which facilitates joint action. Action prediction will be simulated and tested in (WP7).

这个项目的目的是调查人类在交接任务中的联合行动协调。对联合行动控制的洞察可以推进混合（即机器人-人）双体联合行动的控制机制。在我们的实验中，我们测量了在系统地改变二组的组成（例如，伙伴的年龄）和物体的属性（WP1-3）时，在移交任务期间的伸手和抓握运动学以及抓握力动力学。高级分析（例如，降维）将用于理解到达和掌握运动学（WP4-6），从而预测动作后果（例如，切换位置，力缩放），从而促进联合行动。动作预测将在（WP7）中进行模拟和测试。