arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2209.15428cs.RO
收藏

PyPose:一个结合物理优化的机器人学习库

PyPose: A Library for Robot Learning with Physics-based Optimization

Carnegie Mellon University(卡内基梅隆大学);State University of New York at Buffalo(纽约州立大学布法罗分校);Massachusetts Institute of Technology(麻省理工学院);Nanyang Technological University(南洋理工大学);ETH Zürich(苏黎世联邦理工学院);Pennsylvania State University(宾夕法尼亚州立大学);Delhi Technological University(德里理工大学);University of Virginia(弗吉尼亚大学);The Chinese University of Hong Kong, Shenzhen(香港中文大学(深圳));University of Massachusetts Amherst(马萨诸塞大学阿默斯特分校);Georgia Institute of Technology(佐治亚理工学院);Wuhan University(武汉大学);University of Michigan, Ann Arbor(密歇根大学安娜堡分校);Fox Chapel Area High School(福克斯查珀尔地区高中);Lexington High School(列克星敦高中);Stanford University(斯坦福大学)
Chen Wang, Dasong Gao, Kuan Xu, Junyi Geng, Yaoyu Hu, Yuheng Qiu, Bowen Li, Fan Yang, Brady Moon, Abhinav Pandey, Aryan, Jiahe Xu, Tianhao Wu, Haonan He, Daning… 展开作者

Chen Wang, Dasong Gao, Kuan Xu, Junyi Geng, Yaoyu Hu, Yuheng Qiu, Bowen Li, Fan Yang, Brady Moon, Abhinav Pandey, Aryan, Jiahe Xu, Tianhao Wu, Haonan He, Daning Huang, Zhongqiang Ren, Shibo Zhao, Taimeng Fu, Pranay Reddy, Xiao Lin, Wenshan Wang, Jingnan Shi, Rajat Talak, Kun Cao, Yi Du, Han Wang, Huai Yu, Shanzhao Wang, Siyu Chen, Ananth Kashyap, Rohan Bandaru, Karthik Dantu, Jiajun Wu, Lihua Xie, Luca Carlone, Marco Hutter, Sebastian Scherer

AI总结:

PyPose结合深度感知模型与物理优化,提供高效的机器人学习库,实现计算速度提升超过10倍,适用于SLAM、规划、控制等任务。

摘要:

深度学习在机器人感知领域取得了显著成功,但其数据驱动的特性在面对不断变化的环境时表现不佳。相比之下,基于物理的优化具有更好的泛化能力,但在复杂任务中由于缺乏高层语义信息和依赖手动参数调优而表现较差。为利用这两个互补的世界,我们提出了PyPose:一个面向机器人、基于PyTorch的库,将深度感知模型与基于物理的优化相结合。PyPose的架构整洁且组织良好,具有命令式风格的接口,并且高效且易于使用,使其能够轻松集成到实际的机器人应用中。此外,它支持任何阶次的李群和李代数的并行计算以及二阶优化器,如信任区域方法。实验表明,PyPose在计算速度上比最先进的库快了超过10倍。为了促进未来研究,我们为机器人学习的几个领域提供了具体的示例,包括SLAM、规划、控制和惯性导航。

英文摘要:

Deep learning has had remarkable success in robotic perception, but its data-centric nature suffers when it comes to generalizing to ever-changing environments. By contrast, physics-based optimization generalizes better, but it does not perform as well in complicated tasks due to the lack of high-level semantic information and reliance on manual parametric tuning. To take advantage of these two complementary worlds, we present PyPose: a robotics-oriented, PyTorch-based library that combines deep perceptual models with physics-based optimization. PyPose's architecture is tidy and well-organized, it has an imperative style interface and is efficient and user-friendly, making it easy to integrate into real-world robotic applications. Besides, it supports parallel computing of any order gradients of Lie groups and Lie algebras and $2^{\text{nd}}$-order optimizers, such as trust region methods. Experiments show that PyPose achieves more than $10\times$ speedup in computation compared to the state-of-the-art libraries. To boost future research, we provide concrete examples for several fields of robot learning, including SLAM, planning, control, and inertial navigation.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑