Active object detection with multistep action prediction using deep q-network

CORC > 沈阳自动化研究所 > 中国科学院沈阳自动化研究所 > 空间自动化技术研究室

	Active object detection with multistep action prediction using deep q-network
	Sun FC(孙富春)2; Han XN(韩小宁)3 ; Liu HP(刘华平)3; Zhang, Xinyu1
刊名	IEEE Transactions on Industrial Informatics
	2019
卷号	15 期号:6 页码:3723-3731
关键词	Active object detection active vision deep Q-learning network (DQN) dueling architecture reinforcement learning
ISSN号	1551-3203
产权排序	1
英文摘要	In recent years, great success has been achieved in visual object detection, which is one of the fundamental tasks in the field of industrial intelligence. Most of existing methods have been proposed to deal with single well-captured still images, while in practical robotic applications, due to nuisances, such as tiny scale, partial view, or occlusion, one still image may not contain enough information for object detection. However, an intelligent robot has the capability to adjust its viewpoint to get better images for detection. Therefore, active object detection becomes a very important perception strategy for intelligent robots. In this paper, by formulating active object detection as a sequential action decision process, a deep reinforcement learning framework is established to resolve it. Furthermore, a novel deep Q-learning network (DQN) with a dueling architecture is proposed, the network has two separate output channels, one predicts action type and the other predicts action range. By combining the two output channels, the action space is explored more efficiently. Several methods are extensively validated and the results show that the proposed one obtains the best results and predicts action in real time.
语种	英语
WOS记录号	WOS:000471725400053
资助机构	National Science Foundation of China and German Research Foundation under Grant NSFC 61621136008/DFG TRR-169, Grant 91848206 and Grant U1613212
内容类型	期刊论文
源URL	[http://ir.sia.cn/handle/173321/24943]
专题	沈阳自动化研究所_空间自动化技术研究室
通讯作者	Liu HP(刘华平)
作者单位	1.State Key Laboratory of Automotive Safety and Energy, Tsinghua University, Beijing, 100084, China 2.Department of Computer Science and Technology, Tsinghua University, Beijing 100084, China 3.State Key Laboratory of Robotics Shenyang Institute of Automation, Chinese Academy of Sciences University of Chinese Academy of Sciences, Institutes for Robotics and Intelligent Manufacturing, Chinese Academy of Sciences, Shenyang 110016, China
推荐引用方式 GB/T 7714	Sun FC,Han XN,Liu HP,et al. Active object detection with multistep action prediction using deep q-network[J]. IEEE Transactions on Industrial Informatics,2019,15(6):3723-3731.
APA	Sun FC,Han XN,Liu HP,&Zhang, Xinyu.(2019).Active object detection with multistep action prediction using deep q-network.IEEE Transactions on Industrial Informatics,15(6),3723-3731.
MLA	Sun FC,et al."Active object detection with multistep action prediction using deep q-network".IEEE Transactions on Industrial Informatics 15.6(2019):3723-3731.