arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2603.16536cs.RO
收藏

Kamino:基于GPU的多体系统大规模并行模拟器,适用于具有挑战性的拓扑结构

Kamino: GPU-based Simulation of Rigid Multi-Body Physics with Kinematic Loops

Disney Research, Zurich, Switzerland(迪士尼研究所以苏黎世,瑞士);NVIDIA, Zurich, Switzerland(NVIDIA,苏黎世,瑞士)

Vassilios Tsounis, Guirec Maloisel, Christian Schumacher, Ruben Grandia, Agon Serifi, David Müller, Chris Amevor, Tobias Widmer, Moritz Bächer

AI总结:

Kamino通过GPU实现异构高耦合机械系统的并行模拟,支持复杂机械系统中的闭环拓扑结构,利用先进约束优化算法提升仿真精度,实现高效并行模拟与强化学习训练。

摘要:

我们提出了Kamino,一种基于GPU的物理求解器,用于大规模并行模拟异构的高耦合机械系统。该求解器使用NVIDIA Warp用Python实现,并整合到Newton框架中,使数据驱动方法如大规模强化学习能够应用于复杂机械系统,如具有强耦合运动和动态约束的机械系统,如运动闭环。后者常被从业者规避,通过将系统拓扑近似为运动树并加入显式闭环约束或所谓的模仿关节。Kamino旨在通过原生支持这些耦合类型来减轻这一负担。这种能力促进了高吞吐量的并行模拟,能够捕捉利用闭环运动链实现机械优势的机械系统的真正性质。此外,Kamino支持异构世界,允许在单个GPU上对结构各异的机器人进行批量模拟。其核心是一种最先进的约束优化算法,通过将约束刚体多体前向动力学转换为非线性互补性问题来计算约束力。这导致了高保真的仿真,能够在不依赖近似模型简化或凸化问题的情况下解决接触动力学。我们展示了在DR Legs(六层运动闭环双足机器人)上使用强化学习策略训练,生成可行的行走策略,同时在单个GPU上模拟4096个并行环境。

英文摘要:

We present Kamino, a GPU-based physics solver for massively parallel simulations of heterogeneous highly-coupled mechanical systems. Implemented in Python using NVIDIA Warp and integrated into the Newton framework, it enables the application of data-driven methods, such as large-scale reinforcement learning, to complex robotic systems that exhibit strongly coupled kinematic and dynamic constraints such as kinematic loops. The latter are often circumvented by approximating the system's topology as a kinematic tree, and incorporating explicit loop-closure constraints or so-called mimic joints. Kamino aims at alleviating this burden by natively supporting these types of couplings. Moreover, Kamino supports heterogeneous worlds, allowing for batched simulation of structurally diverse robots on a single GPU. At its core lies a state-of-the-art constrained optimization algorithm based on ADMM, which computes constraint forces by solving the constrained rigid multi-body forward dynamics transcribed as a nonlinear complementarity problem. This leads to high-fidelity simulations that resolve contact dynamics without resorting to approximate models that simplify or convexify the problem. We demonstrate RL policy training on a biped with six nested kinematic loops, generating a feasible walking policy while simulating $4{,}096$ parallel environments on a single GPU.

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。

↑