Data61/CSIRO(澳大利亚联邦科学与工业研究组织数据61);Queensland University of Technology (QUT)(昆士兰科技大学);Australian Centre for Robotic Vision (ACRV)(澳大利亚机器人视觉中心);LYRO Robotics Pty Ltd(LYRO机器人私人有限公司)
Jack Collins, Ross Brown, Jurgen Leitner, David Howard
The large demand for simulated data has made the reality gap a problem on the forefront of robotics. We propose a method to traverse the gap by tuning available simulation parameters. Through the optimisation of physics engine parameters, we show that we are able to narrow the gap between simulated solutions and a real world dataset, and thus allow more ready transfer of leaned behaviours between the two. We subsequently gain understanding as to the importance of specific simulator parameters, which is of broad interest to the robotic machine learning community. We find that even optimised for different tasks that different physics engine perform better in certain scenarios and that friction and maximum actuator velocity are tightly bounded parameters that greatly impact the transference of simulated solutions.
CLASH: Collision Learning via Augmented Sim-to-real Hybridization to Bridge the Reality Gap
School of Mathematical Sciences, Peking University(北京大学数学科学学院);School of Artificial Intelligence, Shanghai Jiao Tong University(上海交通大学人工智能学院);School of Mathematical Sciences, Beijing Normal University(北京师范大学数学科学学院)
DROID: Minimizing the Reality Gap using Single-Shot Human Demonstration
Hamlyn Centre for Robotic Surgery(哈姆林 robotic 外科中心);Imperial College London(伦敦帝国学院);School of Computer Science and Engineering, University of Electronic Science and Technology of China(电子科技大学计算机科学与工程学院);Tencent Robotics X(腾讯Robotics X实验室)
Crossing the Reality Gap: a Short Introduction to the Transferability Approach
Université Pierre et Marie Curie-Paris 6(皮埃尔和玛丽·居里大学-巴黎第六大学);CNRS UMR 7222(法国国家科学研究中心UMR 7222实验室)
在机器人技术中,无梯度优化算法(例如进化算法)通常只用于仿真,因为它们需要评估许多候选解。然而,在仿真中获得的解决方案往往不能很好地工作在真实设备上。可移植性方法旨在通过\emph{making the optimization algorithm aware of the limits of the simulation}跨越仿真和现实之间的这一差距。在本文中,我们首先描述的可转移性功能,解决方案的描述符映射到一个分数代表如何以及模拟器匹配的现实。然后,我们表明,这个功能可以使用回归算法和一些实验与真实的设备。我们的研究结果是支持一个简单的四足机器人,其控制参数进行了优化的现实差距的广泛研究。特别是,我们在现实和模拟中映射了整个搜索空间,以了解适应度景观之间的差异。
Robot Learning from Randomized Simulations: A Review
Technical University of Darmstadt(达姆施塔特工业大学);Honda Research Institute Europe(本田欧洲研究所);University of Sydney(悉尼大学);NVIDIA(英伟达);Georgia Institute of Technology(佐治亚理工学院);Robotics at Google(谷歌机器人部门)
Green Robotic Mixed Reality with Gaussian Splatting
Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences(深圳先进技术研究院,中国科学院);University of Chinese Academy of Sciences(中国科学院大学);State Key Laboratory of IOTSC, Department of Computer and Information Science, University of Macau(物联网科学与技术国家重点实验室,澳门大学计算机与信息科学学院);Peng Cheng Laboratory(鹏城实验室);Manifold Tech Limited(曼彻斯特科技有限公司);School of Electrical Engineering and Telecommunications, the University of New South Wales(新南威尔士大学电子与通信工程学院)