A Novel Obstacle Traversal Method for Multiple Robotic Fish Based on Cross-Modal Variational Autoencoders and Imitation Learning

Precision control of multiple robotic fish visual navigation in complex underwater environments has long been a challenging issue in the field of underwater robotics. To address this problem, this paper proposes a multi-robot fish obstacle traversal technique based on the combination of cross-modal...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Biomimetics (Basel, Switzerland) Jg. 9; H. 4; S. 221
Hauptverfasser: Wang, Ruilong, Wang, Ming, Zhao, Qianchuan, Gong, Yanling, Zuo, Lingchen, Zheng, Xuehan, Gao, He
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Switzerland MDPI AG 01.04.2024
Schlagworte:
Algorithms
CM-VAE
Deep learning
Design and construction
Electronics in navigation
Engineering research
Fishes
Image processing
imitation learning
Localization
Machine learning
Mapping
Methods
multiple robotic fish
Navigation systems
Neural networks
Observational learning
Robotics
Robots, Industrial
Sensors
Sensory integration
Underwater navigation
Underwater robots
visual navigation
China
CM-VAE
imitation learning
visual navigation
multiple robotic fish
ISSN:2313-7673, 2313-7673
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Precision control of multiple robotic fish visual navigation in complex underwater environments has long been a challenging issue in the field of underwater robotics. To address this problem, this paper proposes a multi-robot fish obstacle traversal technique based on the combination of cross-modal variational autoencoder (CM-VAE) and imitation learning. Firstly, the overall framework of the robotic fish control system is introduced, where the first-person view of the robotic fish is encoded into a low-dimensional latent space using CM-VAE, and then different latent features in the space are mapped to the velocity commands of the robotic fish through imitation learning. Finally, to validate the effectiveness of the proposed method, experiments are conducted on linear, S-shaped, and circular gate frame trajectories with both single and multiple robotic fish. Analysis reveals that the visual navigation method proposed in this paper can stably traverse various types of gate frame trajectories. Compared to end-to-end learning and purely unsupervised image reconstruction, the proposed control strategy demonstrates superior performance, offering a new solution for the intelligent navigation of robotic fish in complex environments.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ISSN:2313-7673
2313-7673
DOI:10.3390/biomimetics9040221