Interday Reliability of Upper-limb Geometric MyoPassivity Map for Physical Human-Robot Interaction
Interday Reliability of Upper-limb Geometric MyoPassivity Map for Physical Human-Robot Interaction
复制标题
DOI：
10.1109/toh.2023.3277453
复制
发表时间：
2023-05
影响因子：
2.9
通讯作者：
Xingyuan Zhou;Peter Paik;Rory O'Keeffe;S. F. Atashzar
中科院分区：
文献类型：
--
作者：
Xingyuan Zhou;Peter Paik;Rory O'Keeffe;S. F. Atashzar
人类生物力学的内在能量行为的价值最近被认识到并在物理人机交互（pHRI）中被开发。作者最近提出了“被动生物力学过剩”的概念，基于非线性控制理论，构建一个用户特定的能量地图。该地图将评估上肢在与机器人互动时吸收动觉能量的行为。将这些知识集成到pHRI稳定器的设计中可以通过释放隐藏的能量库来减少控制的保守性，从而指示不太保守的稳定裕度。其结果将提高系统的性能，如渲染（遥）触觉系统的动觉透明度。然而，目前的方法需要一个离线的数据驱动的识别程序之前，每一个操作，以估计人体生物力学的能量图。这可能是耗时的，并且挑战易疲劳的用户。在这项研究中，我们首次在五名健康受试者的样本中调查了上肢被动图的日间可靠性。我们的统计分析表明，所确定的被动地图是高度可靠的，在估计预期的充满活力的行为的基础上的组内相关系数分析（在不同的日子和各种相互作用进行）。结果表明，一次性估计是在生物力学感知的pHRI稳定中重复使用的可靠措施，增强了现实生活场景中的实用性。
The value of intrinsic energetic behavior of human biomechanics has recently been recognized and exploited in physical human-robot interaction (pHRI). The authors have recently proposed the concept of “Biomechanical Excess of Passivity,” based on nonlinear control theory, to construct a user-specific energetic map. The map would assess the behavior of the upper-limb in absorbing the kinesthetic energy when interacting with robots. Integrating such knowledge into the design of pHRI stabilizers can reduce the conservatism of the control by unleashing hidden energy reservoirs indicating a less conservative margin of stability. The outcome would enhance the system's performance, such as rendering kinesthetic transparency of (tele)haptics systems. However, current methods require an offline data-driven identification procedure prior to each operation to estimate the energetic map of human biomechanics. This can be time-consuming and challenge users susceptible to fatigue. In this study, for the first time, we investigate the interday reliability of upper-limb passivity maps in a sample of five healthy subjects. Our statistical analyses indicate that the identified passivity map is highly reliable in estimating the expected energetic behavior based on Intraclass correlation coefficient analysis (conducted on different days and with various interactions). The results illustrate that a one-shot estimate is a reliable measure to be used repeatedly in biomechanics-aware pHRI stabilization, enhancing practicality in real-life scenarios.