Echo-Enhanced Embodied Visual Navigation
Neural Computation · 18.04.2023 · 35 (5) · 958-976
DOI: 10.1162/neco_a_01579 ↗Kurz gefasst
Fachartikel in der Zeitschrift Neural Computation, online erschienen am 18.04.2023. Verfasst von Yinfeng Yu und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computer Science and Technology, Tsinghua University, Beijing 100084, China. Die Arbeit wurde bisher 21-mal zitiert (Crossref) und nennt 42 Literaturangaben.
Abstract (Originalsprache)
Visual navigation involves a movable robotic agent striving to reach a point goal (target location) using vision sensory input. While navigation with ideal visibility has seen plenty of success, it becomes challenging in suboptimal visual conditions like poor illumination, where traditional approaches suffer from severe performance degradation. We propose E3VN (echo-enhanced embodied visual navigation) to effectively perceive the surroundings even under poor visibility to mitigate this problem. This is made possible by adopting an echoer that actively perceives the environment via auditory signals. E3VN models the robot agent as playing a cooperative Markov game with that echoer. The action policies of robot and echoer are jointly optimized to maximize the reward in a two-stream actor-critic architecture. During optimization, the reward is also adaptively decomposed into the robot and echoer parts. Our experiments and ablation studies show that E3VN is consistently effective and robust in point goal navigation tasks, especially under nonideal visibility.
Autor:innen und Institutionen
- Department of Computer Science and Technology, Tsinghua University, Beijing 100084, China.
- College of Information Science and Engineering, Xinjiang University, Urumqi 830046, China yyf17@mails.tsinghua.edu.cn.
- Motherbrain, EQT, Stockholm 11153, Sweden lele.cao@eqtpartners.com.
- Department of Computer Science and Technology, Tsinghua University, Beijing 100084, China fcsun@mail.tsinghua.edu.cn.
- Shanghai AI Laboratory, Shanghai 200232, China yangchao@pjlab.org.cn.
- College of Information Science and Engineering, Xinjiang University, Urumqi 830046, China lai@xju.edu.cn.
Zitieren
APA
Yu, Y., Cao, L., Sun, F., Yang, C., Lai, H., & Huang, W. (2023). Echo-Enhanced Embodied Visual Navigation. Neural Computation, 35(5), 958-976. https://doi.org/10.1162/neco_a_01579
Vancouver
Yu Y, Cao L, Sun F, Yang C, Lai H, Huang W. Echo-Enhanced Embodied Visual Navigation. Neural Computation. 2023;35(5):958-976. doi:10.1162/neco_a_01579
BibTeX
@article{Yu2023,
author = {Yu, Yinfeng and Cao, Lele and Sun, Fuchun and Yang, Chao and Lai, Huicheng and Huang, Wenbing},
title = {Echo-Enhanced Embodied Visual Navigation},
journal = {Neural Computation},
year = {2023},
volume = {35},
number = {5},
pages = {958--976},
doi = {10.1162/neco_a_01579}
}
EUVIMED