Instance-Level Knowledge Transfer for Data-Driven Driver Model Adaptation With Homogeneous Domains

doi:10.1109/TITS.2022.3161939

CORC > 自动化研究所 > 中国科学院自动化研究所 > 复杂系统管理与控制国家重点实验室 > 先进控制与自动化团队

	Instance-Level Knowledge Transfer for Data-Driven Driver Model Adaptation With Homogeneous Domains
	Lu, Chao 3; Lv, Chen 2,4; Gong, Jianwei 3; Wang, Wenshuo 1; Cao, Dongpu 6; Wang, Fei-Yue 5
刊名	IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
	2022-03-31
页码	12
关键词	Vehicles Adaptation models Data models Hidden Markov models Knowledge transfer Transfer learning Training Driver behaviour driver model adaptation transfer learning importance weight
ISSN号	1524-9050
DOI	10.1109/TITS.2022.3161939
通讯作者	Gong, Jianwei(gongjianwei@bit.edu.cn)
英文摘要	Driver model adaptation (DMA) plays an essential role for driving behaviour modelling when there is a lack of sufficient data for training the new model. A new data-driven DMA method is proposed in this paper to realise the instance-level knowledge transfer between individual drivers. Using the importance-weighted transfer learning (IWTL), the data collected from one driver (source driver) can be directly used to train the model of another driver (target driver). Under the framework of IWTL, the relationship between two different drivers can be modelled by the importance weight (IW). Two estimation methods Kullback-Leibler (KL) Divergence and least-squares (LS), are used to estimate IW for each data instance by modelling the importance-weight function as a radial basis function (RBF). Experiments based on the driving simulator and real vehicle are carried out to test the performance of TL for steering behaviour adaptation during the overtaking manoeuvre. The experimental results show that the TL method can transfer the knowledge observed from one driver to another when training the new driver model without sufficient data by keeping the modelling error at a low level.
资助项目	National Natural Science Foundation of China[61703041] ; National Natural Science Foundation of China[U19A2083] ; Technological Innovation Program of the Beijing Institute of Technology (BIT)
WOS关键词	STEERING MODEL ; BEHAVIOR ; RECOGNITION ; OVERTAKING ; ASSISTANCE ; VEHICLES
WOS研究方向	Engineering ; Transportation
语种	英语
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
WOS记录号	WOS:000777297900001
资助机构	National Natural Science Foundation of China ; Technological Innovation Program of the Beijing Institute of Technology (BIT)
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/48277]
专题	自动化研究所_复杂系统管理与控制国家重点实验室_先进控制与自动化团队
通讯作者	Gong, Jianwei
作者单位	1.McGill Univ, Dept Civil Engn, Montreal, PQ H3A 0C3, Canada 2.Nanyang Technol Univ, Sch Elect & Elect Engn, Singapore 639798, Singapore 3.Beijing Inst Technol, Sch Mech Engn, Beijing 100081, Peoples R China 4.Nanyang Technol Univ, Sch Mech & Aerosp Engn, Singapore 639798, Singapore 5.Chinese Acad Sci, Inst Automat, Beijing 100190, Peoples R China 6.Tsinghua Univ, Sch Vehicle & Mobil, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Lu, Chao,Lv, Chen,Gong, Jianwei,et al. Instance-Level Knowledge Transfer for Data-Driven Driver Model Adaptation With Homogeneous Domains[J]. IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS,2022:12.
APA	Lu, Chao,Lv, Chen,Gong, Jianwei,Wang, Wenshuo,Cao, Dongpu,&Wang, Fei-Yue.(2022).Instance-Level Knowledge Transfer for Data-Driven Driver Model Adaptation With Homogeneous Domains.IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS,12.
MLA	Lu, Chao,et al."Instance-Level Knowledge Transfer for Data-Driven Driver Model Adaptation With Homogeneous Domains".IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS (2022):12.