Visual-based Autonomous Driving Deployment from a Stochastic and Uncertainty-aware Perspective

被引:0
|
作者
Tai, Lei [1 ]
Yun, Peng [1 ]
Chen, Yuying [1 ]
Liu, Congcong [1 ]
Ye, Haoyang [1 ]
Liu, Ming [1 ]
机构
[1] Hong Kong Univ Sci & Technol, Hong Kong, Peoples R China
来源
2019 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS (IROS) | 2019年
基金
中国国家自然科学基金;
关键词
D O I
10.1109/iros40897.2019.8968307
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
End-to-end visual-based imitation learning has been widely applied in autonomous driving. When deploying the trained visual-based driving policy, a deterministic command is usually directly applied without considering the uncertainty of the input data. Such kind of policies may bring dramatical damage when applied in the real world. In this paper, we follow the recent real-to-sim pipeline by translating the testing world image back to the training domain when using the trained policy. In the translating process, a stochastic generator is used to generate various images stylized under the training domain randomly or directionally. Based on those translated images, the trained uncertainty-aware imitation learning policy would output both the predicted action and the data uncertainty motivated by the aleatoric loss function. Through the uncertainty-aware imitation learning policy, we can easily choose the safest one with the lowest uncertainty among the generated images. Experiments in the Carla navigation benchmark show that our strategy outperforms previous methods, especially in dynamic environments.
引用
收藏
页码:2622 / 2628
页数:7
相关论文
共 50 条
  • [1] Learning an Uncertainty-Aware Object Detector for Autonomous Driving
    Meyer, Gregory P.
    Thakurdesai, Niranjan
    2020 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS (IROS), 2020, : 10521 - 10527
  • [2] Driving Environment Uncertainty-Aware Motion Planning for Autonomous Vehicles
    Xiaolin Tang
    Kai Yang
    Hong Wang
    Wenhao Yu
    Xin Yang
    Teng Liu
    Jun Li
    Chinese Journal of Mechanical Engineering, 2022, 35
  • [3] Learning Effectively from Intervention for Visual-based Autonomous Driving
    Deng, Yunfu
    Xu, Kun
    Hu, Yue
    Cui, Yunduan
    Xiang, Gengzhao
    Pan, Zhongming
    2022 IEEE 25TH INTERNATIONAL CONFERENCE ON INTELLIGENT TRANSPORTATION SYSTEMS (ITSC), 2022, : 4290 - 4295
  • [4] Driving Environment Uncertainty-Aware Motion Planning for Autonomous Vehicles
    Tang, Xiaolin
    Yang, Kai
    Wang, Hong
    Yu, Wenhao
    Yang, Xin
    Liu, Teng
    Li, Jun
    CHINESE JOURNAL OF MECHANICAL ENGINEERING, 2022, 35 (01)
  • [5] Driving Environment Uncertainty-Aware Motion Planning for Autonomous Vehicles
    Xiaolin Tang
    Kai Yang
    Hong Wang
    Wenhao Yu
    Xin Yang
    Teng Liu
    Jun Li
    Chinese Journal of Mechanical Engineering, 2022, 35 (05) : 317 - 330
  • [6] Uncertainty-Aware Model-Based Reinforcement Learning: Methodology and Application in Autonomous Driving
    Wu, Jingda
    Huang, Zhiyu
    Lv, Chen
    IEEE TRANSACTIONS ON INTELLIGENT VEHICLES, 2023, 8 (01): : 194 - 203
  • [7] Uncertainty-Aware Decision-Making for Autonomous Driving at Uncontrolled Intersections
    Tang, Xiaolin
    Zhong, Guichuan
    Li, Shen
    Yang, Kai
    Shu, Keqi
    Cao, Dongpu
    Lin, Xianke
    IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS, 2023, 24 (09) : 9725 - 9735
  • [8] Uncertainty-Aware Sensor Network Deployment
    Reda, Senouci Mustapha
    Abdelhamid, Mellouk
    Latifa, Oukhellou
    Amar, Aissani
    2011 IEEE GLOBAL TELECOMMUNICATIONS CONFERENCE (GLOBECOM 2011), 2011,
  • [9] Efficient Uncertainty-Aware Collision Avoidance for Autonomous Driving Using Convolutions
    Zhang, Chaojie
    Wu, Xichao
    Wang, Jun
    Song, Mengxuan
    IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS, 2024, 25 (10) : 13805 - 13819
  • [10] CCTR: Calibrating Trajectory Prediction for Uncertainty-Aware Motion Planning in Autonomous Driving
    Cao, Chengtai
    Chen, Xinhong
    Wang, Jianping
    Song, Qun
    Tan, Rui
    Li, Yung-Hui
    THIRTY-EIGHTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, VOL 38 NO 19, 2024, : 20949 - 20957