Probabilistic Inference of Simulation Parameters via Parallel Differentiable Simulation
Probabilistic Inference of Simulation Parameters via Parallel Differentiable Simulation
复制标题
DOI：
10.1109/icra46639.2022.9812293
复制
发表时间：
2021-09
期刊：
影响因子：
--
通讯作者：
Eric Heiden;Chris Denniston;David Millard;Fabio Ramos;G. Sukhatme
中科院分区：
文献类型：
--
作者：
Eric Heiden;Chris Denniston;David Millard;Fabio Ramos;G. Sukhatme
在仿真中再现真实世界的动态对于开发新的控制和感知方法至关重要。该任务通常涉及通过以多模态和偏态分布为特征的逆推理问题，从观测到的rollots中估计模拟参数分布。我们通过一种新颖的贝叶斯推理方法来解决这个具有挑战性的问题，该方法近似于给定真实传感器测量的模拟参数的后验分布。通过多次射击公式扩展常用的高斯似然模型的轨迹，我们基于梯度的粒子推理算法，Stein变分梯度下降，能够识别高度非线性，欠驱动的系统。我们利用GPU代码生成和可微模拟来并行评估许多粒子的似然及其梯度。我们的算法比可比基线更准确地推断模拟参数上的非参数分布，并通过基于梯度的优化有效地处理参数上的约束。我们在几个物理实验中评估了估计的性能。在欠驱动机构中，7自由度机械臂激励具有未知质量配置的物体，我们演示了推理技术如何识别参数之间的对称性并提供高度准确的预测。网站:https://uscresl.github.io/prob-diff-sim
Reproducing real world dynamics in simulation is critical for the development of new control and perception methods. This task typically involves the estimation of simu-lation parameter distributions from observed rollouts through an inverse inference problem characterized by multi-modality and skewed distributions. We address this challenging problem through a novel Bayesian inference approach that approximates a posterior distribution over simulation parameters given real sensor measurements. By extending the commonly used Gaus-sian likelihood model for trajectories via the multiple-shooting formulation, our gradient-based particle inference algorithm, Stein Variational Gradient Descent, is able to identify highly nonlinear, underactuated systems. We leverage GPU code gen-eration and differentiable simulation to evaluate the likelihood and its gradient for many particles in parallel. Our algorithm infers nonparametric distributions over simulation parame-ters more accurately than comparable baselines and handles constraints over parameters efficiently through gradient-based optimization. We evaluate estimation performance on several physical experiments. On an underactuated mechanism where a 7-DOF robot arm excites an object with an unknown mass configuration, we demonstrate how the inference technique can identify symmetries between the parameters and provide highly accurate predictions. Website: https://uscresl.github.io/prob-diff-sim