2.5DHANDS: a Gesture-Based MR Remote Collaborative Platform
Peng Wang,Shusheng Zhang,Xiaoliang Bai,Mark Billinghurst,Weiping He,Mengmeng Sun,Yongxing Chen,Hao Lv,Hongyu Ji
DOI: https://doi.org/10.1007/s00170-018-03237-1
IF: 3.563
2019-01-01
The International Journal of Advanced Manufacturing Technology
Abstract:Current remote collaborative systems in manufacturing are mainly based on video-conferencing technology. Their primary aim is to transmit manufacturing process knowledge between remote experts and local workers. However, it does not provide the experts with the same hands-on experience as when synergistically working on site in person. The mixed reality (MR) and increasing networking performances have the capacity to enhance the experience and communication between collaborators in geographically distributed locations. In this paper, therefore, we propose a new gesture-based remote collaborative platform using MR technology that enables a remote expert to collaborate with local workers on physical tasks. Besides, we concentrate on collaborative remote assembly as an illustrative use case. The key advantage compared to other remote collaborative MR interfaces is that it projects the remote expert’s gestures into the real worksite to improve the performance, co-presence awareness, and user collaboration experience. We aim to study the effects of sharing the remote expert’s gestures in remote collaboration using a projector-based MR system in manufacturing. Furthermore, we show the capabilities of our framework on a prototype consisting of a VR HMD, Leap Motion, and a projector. The prototype system was evaluated with a pilot study comparing with the POINTER (adding AR annotations on the task space view through the mouse), which is the most popular method used to augment remote collaboration at present. The assessment adopts the following aspects: the performance, user’s satisfaction, and the user-perceived collaboration quality in terms of the interaction and cooperation. Our results demonstrate a clear difference between the POINTER and 2.5DHANDS interface in the performance time. Additionally, the 2.5DHANDS interface was statistically significantly higher than the POINTER interface in terms of the awareness of user’s attention, manipulation, self-confidence, and co-presence.