Multi-task Learning with Attention for End-to-end Autonomous Driving

被引:21
|
作者
Ishihara, Keishi [1 ,2 ]
Kanervisto, Anssi [2 ]
Miura, Jun [1 ]
Hautamaki, Ville [2 ]
机构
[1] Toyohashi Univ Technol, Toyohashi, Aichi, Japan
[2] Univ Eastern Finland, Kuopio, Finland
关键词
D O I
10.1109/CVPRW53098.2021.00325
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Autonomous driving systems need to handle complex scenarios such as lane following, avoiding collisions, taking turns, and responding to traffic signals. In recent years, approaches based on end-to-end behavioral cloning have demonstrated remarkable performance in point-to-point navigational scenarios, using a realistic simulator and standard benchmarks. Offline imitation learning is readily available, as it does not require expensive hand annotation or interaction with the target environment, but it is difficult to obtain a reliable system. In addition, existing methods have not specifically addressed the learning of reaction for traffic lights, which are a rare occurrence in the training datasets. Inspired by the previous work on multi-task learning and attention modeling, we propose a novel multi-task attention-aware network in the conditional imitation learning (CIL) framework. This does not only improve the success rate of standard benchmarks, but also the ability to react to traffic lights, which we show with standard benchmarks.
引用
收藏
页码:2896 / 2905
页数:10
相关论文
共 50 条
  • [1] End-to-End Multi-Task Learning with Attention
    Liu, Shikun
    Johns, Edward
    Davison, Andrew J.
    [J]. 2019 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2019), 2019, : 1871 - 1880
  • [2] END-TO-END MULTI-TASK MACHINE LEARNING OF VEHICLE DYNAMICS FOR STEERING ANGLE PREDICTION FOR AUTONOMOUS DRIVING
    Merrill, Nicholas
    Eskandarian, Azim
    [J]. PROCEEDINGS OF THE ASME INTERNATIONAL DESIGN ENGINEERING TECHNICAL CONFERENCES AND COMPUTERS AND INFORMATION IN ENGINEERING CONFERENCE, 2019, VOL 3, 2020,
  • [3] An End-to-end Multi-task Object Detection using Embedded GPU in Autonomous Driving
    Zhou, Shanglin
    Xie, Mimi
    Jin, Yufang
    Miao, Fei
    Ding, Caiwen
    [J]. PROCEEDINGS OF THE 2021 TWENTY SECOND INTERNATIONAL SYMPOSIUM ON QUALITY ELECTRONIC DESIGN (ISQED 2021), 2021, : 122 - 128
  • [4] Explaining Autonomous Driving by Learning End-to-End Visual Attention
    Cultrera, Luca
    Seidenari, Lorenzo
    Becattini, Federico
    Pala, Pietro
    Del Bimbo, Alberto
    [J]. 2020 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION WORKSHOPS (CVPRW 2020), 2020, : 1389 - 1398
  • [5] Knowledge Distillation from Single-Task Teachers to Multi-Task Student for End-to-End Autonomous Driving
    Agand, Pedram
    [J]. THIRTY-EIGTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, VOL 38 NO 21, 2024, : 23375 - 23376
  • [6] ATTENTION-AUGMENTED END-TO-END MULTI-TASK LEARNING FOR EMOTION PREDICTION FROM SPEECH
    Zhang, Zixing
    Wu, Bingwen
    Schuller, Bjoern
    [J]. 2019 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP), 2019, : 6705 - 6709
  • [7] Multi-Task End-to-End Self-Driving Architecture for CAV Platoons
    Huch, Sebastian
    Ongel, Aybike
    Betz, Johannes
    Lienkamp, Markus
    [J]. SENSORS, 2021, 21 (04) : 1 - 20
  • [8] End-to-end Spatiotemporal Attention Model for Autonomous Driving
    Zhao, Ruijie
    Zhang, Yanxin
    Huang, Zhiqing
    Yin, Chenkun
    [J]. PROCEEDINGS OF 2020 IEEE 4TH INFORMATION TECHNOLOGY, NETWORKING, ELECTRONIC AND AUTOMATION CONTROL CONFERENCE (ITNEC 2020), 2020, : 2649 - 2653
  • [9] An End-to-End Scalable Iterative Sequence Tagging with Multi-Task Learning
    Gui, Lin
    Du, Jiachen
    Zhao, Zhishan
    He, Yulan
    Xu, Ruifeng
    Fan, Chuang
    [J]. NATURAL LANGUAGE PROCESSING AND CHINESE COMPUTING, NLPCC 2018, PT II, 2018, 11109 : 288 - 298
  • [10] Hybrid Multi-Task Learning for End-To-End Multimodal Emotion Recognition
    Chen, Junjie
    Li, Yongwei
    Zhao, Ziping
    Liu, Xuefei
    Wen, Zhengqi
    Tao, Jianhua
    [J]. 2023 ASIA PACIFIC SIGNAL AND INFORMATION PROCESSING ASSOCIATION ANNUAL SUMMIT AND CONFERENCE, APSIPA ASC, 2023, : 1966 - 1971