Human-Humanoid Robot Interaction through Virtual Reality Interfaces

Human-Humanoid Robot Interaction through Virtual Reality Interfaces
复制标题

DOI:
10.1109/aero50100.2021.9438400
复制
发表时间:
2021-03
期刊:
2021 IEEE Aerospace Conference (50100)
影响因子:
--
通讯作者:
Murphy Wonsick;T. Padır
Murphy Wonsick;T. Padır
中科院分区:
其他
文献类型:
--
作者:
Murphy Wonsick;T. Padır

文献摘要

被引文献

相似文献

在过去的十年中,虚拟现实(VR)取得了巨大的进步,使其成为日常消费者的可用性。随着VR变得越来越普遍,有机会利用这项技术创建直观的操作界面,用于与复杂的动态系统(如人形机器人)进行交互。正如DARPA机器人挑战赛(DRC)所证明的那样,目前的类人机器人界面主要使用带有显示器、键盘和鼠标的标准计算机设置,要求操作员使用2D设备处理3D数据。虽然这些接口可以很好地操作机器人,但它们通常很复杂,需要专业的操作员以及广泛的培训。然而,这种模式可以通过VR改变,允许操作员在3D环境中可视化3D数据并与之交互,从而实现更自然的交互。在本文中,我们介绍了将NASA的人形机器人Valkyrie的典型界面转换为虚拟现实界面的工作。我们比较了《Valkyrie》的标准计算机界面和VR界面,以及使我们的界面成为可能的共享控制计划和系统架构。这项工作的目标是更好地理解虚拟现实界面的效用,以及它们如何在人类监督的机器人应用中使用,以便我们可以朝着更直观和易于使用的控制和交互界面迈进。
In the last decade, there have been great advancements in virtual reality (VR) resulting in its availability for everyday consumers. As VR becomes more ubiquitous, there is an opportunity to utilize this technology to create intuitive operator interfaces for interaction with complex dynamic systems, such as humanoid robots. As evidenced in the DARPA Robotics Challenge (DRC), current interfaces for humanoids primarily use a standard computer setup with monitor, keyboard, and mouse requiring operators to process 3D data with 2D devices. And although these interfaces can be very capable in operating a robot, they are often complex and require expert operators as well as extensive training. However, this paradigm can be changed with VR by allowing operators to visualize and interact with 3D data in a 3D environment, allowing for a more natural interaction. In this paper, we present our work on converting a typical interface to a virtual reality interface for NASA's humanoid robot, Valkyrie. We compare our standard computer interface and our VR interface for Valkyrie, as well as the shared control planners and system architecture that make our interfaces possible. The goal of this work is to better understand the utility of virtual reality interfaces and how they can be employed in human-supervised robot applications so that we may move towards more intuitive and easy-to-use interfaces for control and interaction.