arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2210.08679cs.RO
收藏

基于观测数据的机器人去偏运动估计因果推理

Causal Inference for De-biasing Motion Estimation from Robotic Observational Data

Expedia Group(亿客行集团);U.S. Army Research Laboratory(美国陆军研究实验室)

Junhong Xu, Kai Yin, Jason M. Gregory, Lantao Liu

摘要:

由于安全问题、人类偏好以及任务或平台限制,在复杂的现实世界场景中收集的机器人数据往往存在偏差。因此,机器人学习从这样的观测数据提出了很大的挑战,准确的参数估计。我们提出了一个原则性的因果推理框架,机器人学习参数的随机运动模型使用观测数据。具体来说,我们利用去偏置功能的潜在结果因果推理框架,逆倾向加权(IPW),和双重鲁棒(DR)的方法,以获得更好的参数估计的机器人的随机运动模型。IPW是一种重新加权的方法,以确保无偏估计,和DR方法进一步结合任何两个估计,以加强无偏的结果,即使这些估计之一是有偏的。然后,我们开发了一个近似的策略迭代算法,使用偏差消除估计的状态转移函数。我们使用仿真和真实世界的实验来验证我们的框架,结果表明,所提出的基于因果推理的导航和控制框架可以正确有效地从有偏的观测数据中学习参数。

英文摘要:

Robot data collected in complex real-world scenarios are often biased due to safety concerns, human preferences, and mission or platform constraints. Consequently, robot learning from such observational data poses great challenges for accurate parameter estimation. We propose a principled causal inference framework for robots to learn the parameters of a stochastic motion model using observational data. Specifically, we leverage the de-biasing functionality of the potential-outcome causal inference framework, the Inverse Propensity Weighting (IPW), and the Doubly Robust (DR) methods, to obtain a better parameter estimation of the robot's stochastic motion model. The IPW is a re-weighting approach to ensure unbiased estimation, and the DR approach further combines any two estimators to strengthen the unbiased result even if one of these estimators is biased. We then develop an approximate policy iteration algorithm using the bias-eliminated estimated state transition function. We validate our framework using both simulation and real-world experiments, and the results have revealed that the proposed causal inference-based navigation and control framework can correctly and efficiently learn the parameters from biased observational data.

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。

↑