VRGym:A Virtual Testbed for Physical and Interactive Al
We propose VRGym,a virtual reality(VR)testbed for realistic human-robot interaction.Different from existing toolkits and VR environments,the VRGym emphasizes on building and training both physical and interactive agents for robotics,machine learning,and cognitive science.VRGym leverages mechanisms that can generate diverse 3D scenes with high realism through physics-based simu-lation.We demonstrate that VRGym is able to(ⅰ)collect human interactions and fine manipulations,(ⅱ)accommodate various robots with a ROS bridge,(ⅲ)support experiments for human-robot in-teraction,and(ⅳ)provide toolkits for training the state-of-the-art machine learning algorithms.We hope VRGym can help to advance general-purpose robotics and machine learning agents,as well as assisting human studies in the field of cognitive science.
Virtual reality simulation training benchmark ROS
Xu Xie Hangxin Liu Zhenliang Zhang Yuxing Qiu Feng Gao Siyuan Qi Yixin Zhu Song-Chun Zhu
UCLA Center for Vision,Cognition,Learning,and Autonomy Los Angeles,CA 90095
国际会议
2019国图灵大会(ACM Turing Celebration conference-China 2019 )
成都
英文
225-230
2019-05-17（万方平台首次上网日期，不代表论文的发表时间）