Adaptive cruise control via adaptive dynamic programming with experience replay

doi:10.1007/s00500-018-3063-7

CASIA OpenIR > 多模态人工智能系统全国重点实验室 > 深度强化学习

	Adaptive cruise control via adaptive dynamic programming with experience replay
	Wang, Bin1,2 ; Zhao, Dongbin2 ; Cheng, Jin 1
发表期刊	SOFT COMPUTING
ISSN	1432-7643
	2019-06-01
卷号	23 期号:12 页码:4131-4144
通讯作者	Wang, Bin(cse_wangb@ujn.edu.cn)
摘要	The adaptive cruise control (ACC) problem can be transformed to an optimal tracking control problem for complex nonlinear systems. In this paper, a novel highly efficient model-free adaptive dynamic programming (ADP) approach with experience replay technology is proposed to design the ACC controller. Experience replay increases the data efficiency by recording the available driving data and repeatedly presenting them to the learning procedure of the acceleration controller in the ACC system. The learning framework that combines ADP with experience replay is described in detail. The distinguishing feature of the algorithm is that when estimating parameters of the critic network and the actor network with gradient rules, the gradients of historical data and current data are used to update parameters concurrently. It is proved with Lyapunov theory that the weight estimation errors of the actor network and the critic network are uniformly ultimately bounded under the novel weight update rules. The learning performance of the ACC controller implemented by this ADP algorithm is clearly demonstrated that experience replay can increase data efficiency significantly, and the approximate optimality and adaptability of the learned control policy are tested with typical driving scenarios.
关键词	Adaptive cruise control Adaptive dynamic programming Experience replay Reinforcement learning Neural networks
DOI	10.1007/s00500-018-3063-7
关键词[WOS]	SYSTEMS ; DESIGN
收录类别	SCI
语种	英语
资助项目	National Natural Science Foundation of China[61603150] ; National Natural Science Foundation of China[61273136] ; National Natural Science Foundation of China[61573353] ; National Natural Science Foundation of China[61533017] ; National Key Research and Development Plan[2016YFB0101000] ; Doctoral Foundation of University of Jinan[XBS1605] ; National Natural Science Foundation of China[61603150] ; National Natural Science Foundation of China[61273136] ; National Natural Science Foundation of China[61573353] ; National Natural Science Foundation of China[61533017] ; National Key Research and Development Plan[2016YFB0101000] ; Doctoral Foundation of University of Jinan[XBS1605] ; National Natural Science Foundation of China[61603150] ; National Natural Science Foundation of China[61273136] ; National Natural Science Foundation of China[61573353] ; National Natural Science Foundation of China[61533017] ; National Key Research and Development Plan[2016YFB0101000] ; Doctoral Foundation of University of Jinan[XBS1605]
项目资助者	National Natural Science Foundation of China ; National Key Research and Development Plan ; Doctoral Foundation of University of Jinan
WOS研究方向	Computer Science
WOS类目	Computer Science, Artificial Intelligence ; Computer Science, Interdisciplinary Applications
WOS记录号	WOS:000466948100018
出版者	SPRINGER
七大方向——子方向分类	强化与进化学习
引用统计	被引频次：8[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/24397
专题	多模态人工智能系统全国重点实验室_深度强化学习
通讯作者	Wang, Bin
作者单位	1.Univ Jinan, Sch Elect Engn, Jinan 250022, Shandong, Peoples R China 2.Chinese Acad Sci, Inst Automat, State Key Lab Management & Control Complex Syst, Beijing 100190, Peoples R China
第一作者单位	中国科学院自动化研究所
通讯作者单位	中国科学院自动化研究所
推荐引用方式 GB/T 7714	Wang, Bin,Zhao, Dongbin,Cheng, Jin. Adaptive cruise control via adaptive dynamic programming with experience replay[J]. SOFT COMPUTING,2019,23(12):4131-4144.
APA	Wang, Bin,Zhao, Dongbin,&Cheng, Jin.(2019).Adaptive cruise control via adaptive dynamic programming with experience replay.SOFT COMPUTING,23(12),4131-4144.
MLA	Wang, Bin,et al."Adaptive cruise control via adaptive dynamic programming with experience replay".SOFT COMPUTING 23.12(2019):4131-4144.