arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2312.00548cs.LGcs.CVcs.RO
收藏

基于视觉观测的域自适应模仿学习

Domain Adaptive Imitation Learning with Visual Observation

KAIST(韩国科学技术院);UNIST(蔚山国立科学技术院);Carnegie Mellon University(卡内基梅隆大学);LG AI Research(LG人工智能研究院)

Sungho Choi, Seungyul Han, Woojun Kim, Jongseong Chae, Whiyoung Jung, Youngchul Sung

AI总结:

本文针对视觉观测下跨域模仿学习中的域偏移问题,提出基于双特征提取与图像重建的域无关行为特征提取框架,实验证明其性能优于已有算法。

摘要:

在本文中,我们考虑基于视觉观测的域自适应模仿学习,其中目标域中的智能体通过观察源域中的专家演示来学习执行任务。域自适应模仿学习出现在实际场景中:一个接收视觉感知数据的机器人,需要通过从不同角度视觉观察其他机器人或观察不同形状的机器人来模仿其运动。为了克服基于视觉观测的跨域模仿学习中的域偏移问题,我们提出了一种新颖的框架,该框架基于双特征提取和图像重建,从输入观测中提取与域无关的行为特征,并利用这些特征来训练学习器。实验结果表明,我们的方法优于以往在存在域偏移情况下基于视觉观测的模仿学习算法。

英文摘要:

In this paper, we consider domain-adaptive imitation learning with visual observation, where an agent in a target domain learns to perform a task by observing expert demonstrations in a source domain. Domain adaptive imitation learning arises in practical scenarios where a robot, receiving visual sensory data, needs to mimic movements by visually observing other robots from different angles or observing robots of different shapes. To overcome the domain shift in cross-domain imitation learning with visual observation, we propose a novel framework for extracting domain-independent behavioral features from input observations that can be used to train the learner, based on dual feature extraction and image reconstruction. Empirical results demonstrate that our approach outperforms previous algorithms for imitation learning from visual observation with domain shift.

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。

↑