DeepNav: Learning to Navigate Large Cities

被引:24
|
作者
Brahmbhatt, Samarth [1 ]
Hays, James [1 ]
机构
[1] Georgia Inst Technol, Atlanta, GA 30332 USA
来源
30TH IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2017) | 2017年
关键词
D O I
10.1109/CVPR.2017.329
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
We present DeepNav, a Convolutional Neural Network (CNN) based algorithm for navigating large cities using locally visible street-view images. The DeepNav agent learns to reach its destination quickly by making the correct navigation decisions at intersections. We collect a large-scale dataset of street-view images organized in a graph where nodes are connected by roads. This dataset contains 10 city graphs and more than 1 million street-view images. We propose 3 supervised learning approaches for the navigation task and show how A* search in the city graph can be used to generate supervision for the learning. Our annotation process is fully automated using publicly available mapping services and requires no human input. We evaluate the proposed DeepNav models on 4 held-out cities for navigating to 5 different types of destinations. Our algorithms outperform previous work that uses hand-crafted features and Support Vector Regression (SVR) [19].
引用
收藏
页码:3087 / 3096
页数:10
相关论文
共 50 条
  • [21] Learning to Navigate Cloth using Haptics
    Clegg, Alexander
    Yu, Wenhao
    Erickson, Zackory
    Tan, Jie
    Liu, C. Karen
    Turk, Greg
    2017 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS (IROS), 2017, : 2799 - 2805
  • [22] Learning to Navigate Through Crowded Environments
    Henry, Peter
    Vollmer, Christian
    Ferris, Brian
    Fox, Dieter
    2010 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION (ICRA), 2010, : 981 - 986
  • [23] Learning to navigate the geography of citation indexes
    Debbie Chaves
    Nature, 2008, 453 : 719 - 719
  • [24] Learning to Navigate Endoscopic Capsule Robots
    Turan, Mehmet
    Almalioglu, Yasin
    Gilbert, Hunter B.
    Mahmood, Faisal
    Durr, Nicholas J.
    Araujo, Helder
    Sari, Alp Eren
    Ajay, Anurag
    Sitti, Metin
    IEEE ROBOTICS AND AUTOMATION LETTERS, 2019, 4 (03) : 3075 - 3082
  • [25] DeepNav: Joint View Learning for Direct Optimal Path Perception in Cochlear Surgical Platform Navigation
    Zamani, Majid
    Demosthenous, Andreas
    IEEE ACCESS, 2023, 11 : 120593 - 120602
  • [26] How teachers navigate their learning in developing inclusive learning communities
    Swart, Estelle
    Oswald, Marietjie
    EDUCATION AS CHANGE, 2008, 12 (02) : 91 - 108
  • [27] Learning to Navigate in Human Environments via Deep Reinforcement Learning
    Gao, Xingyuan
    Sun, Shiying
    Zhao, Xiaoguang
    Tan, Min
    NEURAL INFORMATION PROCESSING (ICONIP 2019), PT I, 2019, 11953 : 418 - 429
  • [28] Learning to Navigate for Fine-Grained Classification
    Yang, Ze
    Luo, Tiange
    Wang, Dong
    Hu, Zhiqiang
    Gao, Jun
    Wang, Liwei
    COMPUTER VISION - ECCV 2018, PT XIV, 2018, 11218 : 438 - 454
  • [29] MLNav: Learning to Safely Navigate on Martian Terrains
    Daftry, Shreyansh
    Abcouwer, Neil
    Del Sesto, Tyler
    Venkatraman, Siddarth
    Song, Jialin
    Igel, Lucas
    Byon, Amos
    Rosolia, Ugo
    Yue, Yisong
    Ono, Masahiro
    IEEE ROBOTICS AND AUTOMATION LETTERS, 2022, 7 (02) : 5461 - 5468
  • [30] Learning to navigate efficiently and precisely in real environments
    Bono, Guillaume
    Poirier, Herve
    Antsfeld, Leonid
    Monaci, Gianluca
    Chidlovskii, Boris
    Wolf, Christian
    2024 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2024, : 17837 - 17846