MBDS
收藏资源简介:
MBDS数据集由中国科学院软件研究所创建,是一个用于多体动力学模拟的高质量数据集。该数据集涵盖了1D、2D和3D场景,包含15000条运动轨迹和200个时间步长,远超现有数据集。数据集的创建过程中,研究人员引入了复杂的机械结构,以更真实地模拟物理世界。MBDS数据集主要应用于图网络模拟器(GNS)的训练和评估,旨在解决现有模拟方法在复杂工业场景中的不足,提升物理系统建模和模拟的准确性。
The MBDS dataset, created by the Institute of Software, Chinese Academy of Sciences, is a high-quality dataset for multibody dynamics simulation. Covering 1D, 2D and 3D scenarios, it contains 15,000 motion trajectories and 200 time steps, which far outperforms existing datasets. During its development, researchers introduced complex mechanical structures to simulate the physical world more realistically. The MBDS dataset is primarily applied to the training and evaluation of Graph Network Simulator (GNS), aiming to address the limitations of current simulation methods in complex industrial scenarios and improve the accuracy of physical system modeling and simulation.
MBDS数据集概述
数据集名称
- MBDS
数据集全称
- MBDS: A Multi-Body Dynamics Simulation Dataset for Graph Networks Simulators
数据集描述
- 该数据集用于多体动力学模拟,适用于图网络模拟器。
数据集状态
- 数据集尚未上传,预计即将上传。

- 1MBDS: A Multi-Body Dynamics Simulation Dataset for Graph Networks Simulators中国科学院软件研究所 · 2024年



