GAPLE: Generalizable Approaching Policy LEarning for Robotic Object Searching in Indoor Environment
GAPLE: Generalizable Approaching Policy LEarning for Robotic Object Searching in Indoor Environment
复制标题
DOI：
10.1109/lra.2019.2930426
复制
发表时间：
2018-09
影响因子：
5.2
通讯作者：
Xin Ye;Zhe L. Lin;Joon-Young Lee;Jianming Zhang;Shibin Zheng;Yezhou Yang
中科院分区：
文献类型：
--
作者：
Xin Ye;Zhe L. Lin;Joon-Young Lee;Jianming Zhang;Shibin Zheng;Yezhou Yang
我们研究了一个智能代理人在室内环境中，仅从其视觉输入中学习智能代理人积极地接触感兴趣的对象的问题的问题。尽管已经广泛研究了以场景为导向或以识别为导向的视觉导航，但先前的努力严重遭受了有限的概括能力。在这封信中，我们首先认为对象搜索任务是环境依赖性的，而接近能力是一般的。为了学习可概括的即将来临的政策，我们提出了一种新颖的解决方案，称为可概括的即将到来的政策学习，该解决方案采用了两个视觉特征渠道：深度和语义细分，作为政策学习模块的输入。在现实情况下，在House3D数据集和物理平台上进行的实证研究验证了我们的假设，我们进一步提供了深入的定性分析。
We study the problem of learning a generalizable action policy for an intelligent agent to actively approach an object of interest, in an indoor environment, solely from its visual inputs. While scene-driven or recognition-driven visual navigation has been widely studied, prior efforts suffer severely from the limited generalization capability. In this letter, we first argue the object searching task is environment-dependent while the approaching ability is general. To learn a generalizable approaching policy, we present a novel solution dubbed as Generalizable Approaching Policy LEarning, which adopts two channels of visual features: depth and semantic segmentation, as the inputs to the policy learning module. The empirical studies conducted on the House3D dataset and on a physical platform in a real-world scenario validate our hypothesis, and we further provide in-depth qualitative analysis.