End-to-End Autonomous Driving: An Angle Branched Network Approach

doi:10.1109/TVT.2019.2921918

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

	End-to-End Autonomous Driving: An Angle Branched Network Approach
	Wang, Qing 7; Chen, Long 7; Tian, Bin 4,5,6; Tian, Wei 3; Li, Lingxi 2; Cao, Dongpu 1
刊名	IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY
	2019-12-01
卷号	68 期号:12 页码:11599-11610
关键词	End-to-end driving imitation learning navigation command
ISSN号	0018-9545
DOI	10.1109/TVT.2019.2921918
通讯作者	Chen, Long(chenl46@mail.sysu.edu.cn)
英文摘要	Imitation learning for the end-to-end autonomous driving has drawn renewed attention from academic communities. Current methods either only use images as the input, which will yield ambiguities when a vehicle approaches an intersection, or use additional command information to navigate the vehicle but inefficiently. Focusing on making the vehicle automatically drive along the given path, we propose a new and effective navigation command called as subgoal anglewhich does not require human participation and is calculated by the current position and subgoal of the ego-vehicle. Thus, the subgoal angle contains more information than the navigation command represented as a one-hot vector. Additionally, we propose a model architecture called as angle branched network that makes predictions based on the subgoal angle. In this network, the subgoal angle is not only used for extracting useful features but also for guiding the appropriate prediction layer to make predictions for both the steer angle and the throttle status (which controls the acceleration). Experiments are conducted in a three-dimensional urban simulator. Both quantitive and qualitive results show the effectiveness of the navigation command and the angle branched network. Moreover, the performance can be further boosted by integrating both semantic and depth information into the driving model. Especially by using the depth information, collisions with vehicles and pedestrians can be reduced.
资助项目	National Key R&D Program of China[2018YFB1305002] ; National Natural Science Foundation of China[61773414]
WOS关键词	VEHICLE NAVIGATION
WOS研究方向	Engineering ; Telecommunications ; Transportation
语种	英语
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
WOS记录号	WOS:000518226900020
资助机构	National Key R&D Program of China ; National Natural Science Foundation of China
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/38373]
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	Chen, Long
作者单位	1.Univ Waterloo, Dept Mech & Mechatron Engn, Waterloo, ON N2L 3G1, Canada 2.Indiana Univ Purdue Univ, Purdue Sch Engn & Technol, Dept Elect & Comp Engn, Indianapolis, IN 46202 USA 3.Karlsruhe Inst Technol, Inst Measurement & Control Syst, D-76131 Karlsruhe, Germany 4.Chinese Acad Sci, Cloud Comp Ctr, Dongguan 523808, Peoples R China 5.Qingdao Acad Intelligent Ind, Qingdao 266111, Shandong, Peoples R China 6.Chinese Acad Sci, Inst Automat, State Key Lab Management & Control Complex Syst, Beijing 100190, Peoples R China 7.Sun Yat Sen Univ, Sch Data & Comp Sci, Guangzhou 510245, Peoples R China
推荐引用方式 GB/T 7714	Wang, Qing,Chen, Long,Tian, Bin,et al. End-to-End Autonomous Driving: An Angle Branched Network Approach[J]. IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY,2019,68(12):11599-11610.
APA	Wang, Qing,Chen, Long,Tian, Bin,Tian, Wei,Li, Lingxi,&Cao, Dongpu.(2019).End-to-End Autonomous Driving: An Angle Branched Network Approach.IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY,68(12),11599-11610.
MLA	Wang, Qing,et al."End-to-End Autonomous Driving: An Angle Branched Network Approach".IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY 68.12(2019):11599-11610.