Data-Free Learning of Reduced-Order Kinematics
Data-Free Learning of Reduced-Order Kinematics
复制标题
DOI：
10.1145/3588432.3591521
复制
发表时间：
2023-05
期刊：
影响因子：
--
通讯作者：
Nicholas Sharp;Cristian Romero;Alec Jacobson;E. Vouga;P. Kry;D. Levin;J. Solomon
中科院分区：
文献类型：
--
作者：
Nicholas Sharp;Cristian Romero;Alec Jacobson;E. Vouga;P. Kry;D. Levin;J. Solomon
从弹性体到运动学连杆的物理系统定义在高维构型空间上，但其典型的低能量构型集中在低得多的维子空间上。这项工作解决了自动识别这样的子空间的挑战：给定高维系统的能量函数作为输入，我们产生一个低维映射，其图像参数化为各种但低能量的构型子流形。唯一需要的额外输入是系统初始化程序的单一种子配置；不需要轨迹数据集。我们将子空间表示为神经网络，将低维潜在向量映射到完整的配置空间，并提出了一种训练方案，以适应任何感兴趣的系统的网络参数。我们的实验不仅展示了非线性和非常低维的弹性体和布子空间，而且还展示了更一般的系统，如碰撞刚体和连杆。我们简要地探索了建立在该公式基础上的应用，包括操纵、潜在内插和采样。
Physical systems ranging from elastic bodies to kinematic linkages are defined on high-dimensional configuration spaces, yet their typical low-energy configurations are concentrated on much lower-dimensional subspaces. This work addresses the challenge of identifying such subspaces automatically: given as input an energy function for a high-dimensional system, we produce a low-dimensional map whose image parameterizes a diverse yet low-energy submanifold of configurations. The only additional input needed is a single seed configuration for the system to initialize our procedure; no dataset of trajectories is required. We represent subspaces as neural networks that map a low-dimensional latent vector to the full configuration space, and propose a training scheme to fit network parameters to any system of interest. This formulation is effective across a very general range of physical systems; our experiments demonstrate not only nonlinear and very low-dimensional elastic body and cloth subspaces, but also more general systems like colliding rigid bodies and linkages. We briefly explore applications built on this formulation, including manipulation, latent interpolation, and sampling.