MQA: Answering the Question via Robotic Manipulation
MQA: Answering the Question via Robotic Manipulation
复制标题
MQA：通过机器人操作回答问题
DOI：
10.15607/rss.2021.xvii.044
复制
发表时间：
2020
期刊：
影响因子：
--
通讯作者：
Jing
中科院分区：
文献类型：
--
作者：
Yuhong Deng;Naifu Zhang;Di Guo;Huaping Liu;F. Sun;Chen Pang;Jing
在本文中，我们提出了一种新的操作问答(MQA)任务，这是一类问答任务，要求机器人通过操作与环境进行主动交互来找到问题的答案。考虑到桌面场景，生成场景的热图以帮助机器人对场景进行语义理解，并提出了一种带有语义理解度量的模仿学习方法来生成操作动作，以指导机械手对桌面进行探索以找到问题的答案。此外，还建立了一个包含各种桌面场景和相应问答对的新型数据集。为了验证该框架的有效性，已经进行了大量的实验。
In this paper,we propose a novel task of Manipulation Question Answering(MQA),a class of Question Answering (QA) task, where the robot is required to find the answer to the question by actively interacting with the environment via manipulation. Considering the tabletop scenario, a heatmap of the scene is generated to facilitate the robot to have a semantic understanding of the scene and an imitation learning approach with semantic understanding metric is proposed to generate manipulation actions which guide the manipulator to explore the tabletop to find the answer to the question. Besides, a novel dataset which contains a variety of tabletop scenarios and corresponding question-answer pairs is established. Extensive experiments have been conducted to validate the effectiveness of the proposed framework.
SHOP-VRB：对象感知的视觉推理基准
DOI：
10.1109/icra40945.2020.9197332
发表时间：
2020
期刊：
--
影响因子：
--
作者：
Nazarczuk M
通讯作者：
Nazarczuk M