Learning Physical Constraints with Neural Projections
Learning Physical Constraints with Neural Projections
复制标题
DOI：
--
复制
发表时间：
2020-06
期刊：
影响因子：
--
通讯作者：
Shuqi Yang;Xingzhe He;Bo Zhu
中科院分区：
文献类型：
--
作者：
Shuqi Yang;Xingzhe He;Bo Zhu
我们提出了一类新的神经网络，通过学习它们的基础约束来预测物理系统的行为。神经投影算子是我们方法的核心，它由一个具有嵌入式递归体系结构的轻量级网络组成，该网络交互地实施学习的支撑约束，并预测不同物理系统的各种治理行为。我们的神经投影算子是由基于位置的动力学模型驱动的，该模型已广泛应用于游戏和视觉效果行业，以统一各种快速物理模拟器。我们的方法可以自动有效地从观测点数据中发现广泛的约束，如长度、角度、弯曲、碰撞、边界效应及其任意组合，而不需要任何连接先验。我们提供多组点表示与可配置的网络连接机制相结合，以结合用于处理复杂物理系统的先前输入。我们以统一而简单的方式学习了一组具有挑战性的物理系统，包括具有复杂几何形状的刚体、具有不同长度和弯曲的绳索、铰接式软体和刚体以及具有复杂边界的多对象碰撞，从而展示了我们方法的有效性。
We propose a new family of neural networks to predict the behaviors of physical systems by learning their underpinning constraints. A neural projection operator liesat the heart of our approach, composed of a lightweight network with an embedded recursive architecture that interactively enforces learned underpinning constraints and predicts the various governed behaviors of different physical systems. Our neural projection operator is motivated by the position-based dynamics model that has been used widely in game and visual effects industries to unify the various fast physics simulators. Our method can automatically and effectively uncover a broad range of constraints from observation point data, such as length, angle, bending, collision, boundary effects, and their arbitrary combinations, without any connectivity priors. We provide a multi-group point representation in conjunction with a configurable network connection mechanism to incorporate prior inputs for processing complex physical systems. We demonstrated the efficacy of our approach by learning a set of challenging physical systems all in a unified and simple fashion including: rigid bodies with complex geometries, ropes with varying length and bending, articulated soft and rigid bodies, and multi-object collisions with complex boundaries.