Abstract:
Humanoid grasping is a critical ability for anthropomorphic hand, and plays a significant role in the development of humanoid robots. In this article, we present a deep learning-based control framework for humanoid grasping, incorporating the dynamic contact process among the anthropomorphic hand, the object, and the environment. This method efficiently eliminates the constraints imposed by inaccessible grasping points on both the contact surface of the object and the table surface. To mimic human-like grasping movements, an underactuated anthropomorphic hand is utilized, which is designed based on human hand data. The utilization of hand gestures, rather than controlling each motor separately, has significantly decreased the control dimensionality. Additionally, a deep learning framework is used to select gestures and grasp actions. Our methodology, proven both in simulation and on real robot, exceeds the performance of static analysis-based methods, as measured by the standard grasp metric Q1. It expands the range of objects the system can handle, effectively grasping thin items such as cards on tables, a task beyond the capabilities of previous methodologies.
摘要:
人形抓握是拟人化手的关键能力,并在人形机器人的发展中起着重要作用。在这篇文章中,我们提出了一个基于深度学习的人形抓取控制框架,结合拟人化手之间的动态接触过程,对象,和环境。该方法有效地消除了由物体的接触表面和桌子表面上的不可接近的抓握点施加的约束。为了模仿人类般的抓握动作,一只驱动不足的拟人化手被利用,这是基于人手数据设计的。手势的利用,而不是单独控制每个电机,显著降低了控制维度。此外,深度学习框架用于选择手势和掌握动作。我们的方法论,在仿真和真实机器人上都得到了证明,超过了基于静态分析的方法的性能,按标准把握度量Q1。它扩展了系统可以处理的对象范围,有效地抓住薄的物品,如桌子上的卡片,一项超越先前方法能力的任务。