Multi-task safe reinforcement learning for navigating intersections in dense traffic

doi:10.1016/j.jfranklin.2022.06.052

CORC > 自动化研究所 > 中国科学院自动化研究所 > 多模态人工智能系统全国重点实验室

	Multi-task safe reinforcement learning for navigating intersections in dense traffic
	Liu, Yuqi 1,2; Gao, Yinfeng 3; Zhang, Qichao 1,2; Ding, Dawei 3; Zhao, Dongbin 1,2
刊名	JOURNAL OF THE FRANKLIN INSTITUTE-ENGINEERING AND APPLIED MATHEMATICS
	2023-11-01
卷号	360 期号:17 页码:13737-13760
ISSN号	0016-0032
DOI	10.1016/j.jfranklin.2022.06.052
通讯作者	Zhang, Qichao(zhangqichao2014@ia.ac.cn)
英文摘要	Multi-task intersection navigation, which includes unprotected turning left, turning right, and going straight in heavy traffic, remains a difficult task for autonomous vehicles. For the human driver, negotiation skills with other interactive vehicles are the key to guaranteeing safety and efficiency. However, it is hard to balance the safety and efficiency of the autonomous vehicle for multi-task intersection navigation. In this paper, we formulate a multi-task safe reinforcement learning framework with social attention to improve the safety and efficiency when interacting with other traffic participants. Specifically, the social attention module is used to focus on the states of negotiation vehicles. In addition, a safety layer is added to the multi-task reinforcement learning framework to guarantee safe negotiation. We deploy experiments in the simulators SUMO, which has abundant traffic flows, and CARLA, which has high-fidelity vehicle models. Both show that the proposed algorithm improves safety while maintaining stable traffic efficiency for the multi-task intersection navigation problem. More details and demonstrations are available at https:// github.com/ liuyuqi123/ SAT.(c) 2022 The Franklin Institute. Published by Elsevier Ltd. All rights reserved.
资助项目	National Natural Science Foundation of China (NSFC)[62136008] ; National Natural Science Foundation of China (NSFC)[62173325] ; Beijing Science and Technology Plan[Z191100007419002]
WOS关键词	MODEL
WOS研究方向	Automation & Control Systems ; Engineering ; Mathematics
语种	英语
出版者	PERGAMON-ELSEVIER SCIENCE LTD
WOS记录号	WOS:001111154800001
资助机构	National Natural Science Foundation of China (NSFC) ; Beijing Science and Technology Plan
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/55110]
专题	多模态人工智能系统全国重点实验室
通讯作者	Zhang, Qichao
作者单位	1.Chinese Acad Sci, Inst Automat, State Key Lab Management & Control Complex Syst, Beijing 100190, Peoples R China 2.Univ Chinese Acad Sci, Coll Artificial Intelligence, Beijing 100049, Peoples R China 3.Univ Sci & Technol Beijing, Sch Automat & Elect Engn, Beijing 1000839, Peoples R China
推荐引用方式 GB/T 7714	Liu, Yuqi,Gao, Yinfeng,Zhang, Qichao,et al. Multi-task safe reinforcement learning for navigating intersections in dense traffic[J]. JOURNAL OF THE FRANKLIN INSTITUTE-ENGINEERING AND APPLIED MATHEMATICS,2023,360(17):13737-13760.
APA	Liu, Yuqi,Gao, Yinfeng,Zhang, Qichao,Ding, Dawei,&Zhao, Dongbin.(2023).Multi-task safe reinforcement learning for navigating intersections in dense traffic.JOURNAL OF THE FRANKLIN INSTITUTE-ENGINEERING AND APPLIED MATHEMATICS,360(17),13737-13760.
MLA	Liu, Yuqi,et al."Multi-task safe reinforcement learning for navigating intersections in dense traffic".JOURNAL OF THE FRANKLIN INSTITUTE-ENGINEERING AND APPLIED MATHEMATICS 360.17(2023):13737-13760.