EUVIMEDEUVIMED entdecken
Crossref

Reinforcement learning–based framework for whale rendezvous via autonomous sensing robots

Ninad Jadhav iD, Sushmita Bhattacharya iD, Daniel Vogt iD, Yaniv Aluma iD, Pernille Tønnesen iD, Akarsh Prabhakara, Swarun Kumar iD, Shane Gero iD, Robert J Wood iD, Stephanie Gil

Science Robotics · 2024

DOI: 10.1126/scirobotics.adn7299 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Science Robotics, 2024. Verfasst von Ninad Jadhav und 9 weiteren Autor:innen. Beteiligte Einrichtung: Project CETI, New York, NY, USA. Hauptthemen laut MEDLINE: Robotics, Algorithms, Sperm Whale, Vocalization, Animal. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 73 Literaturangaben.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Rendezvous with sperm whales for biological observations is made challenging by their prolonged dive patterns. Here, we propose an algorithmic framework that codevelops multiagent reinforcement learning–based routing (autonomy module) and synthetic aperture radar–based very high frequency (VHF) signal–based bearing estimation (sensing module) for maximizing rendezvous opportunities of autonomous robots with sperm whales. The sensing module is compatible with low-energy VHF tags commonly used for tracking wildlife. The autonomy module leverages in situ noisy bearing measurements of whale vocalizations, VHF tags, and whale dive behaviors to enable time-critical rendezvous of a robot team with multiple whales in simulation. We conducted experiments at sea in the native habitat of sperm whales using an “engineered whale”—a speedboat equipped with a VHF-emitting tag, emulating five distinct whale tracks, with different whale motions. The sensing module shows a median bearing error of 10.55° to the tag. Using bearing measurements to the engineered whale from an acoustic sensor and our sensing module, our autonomy module gives an aggregate rendezvous success rate of 81.31% for a 500-meter rendezvous distance using three robots in postprocessing. A second class of fielded experiments that used acoustic-only bearing measurements to three untagged sperm whales showed an aggregate rendezvous success rate of 68.68% for a 1000-meter rendezvous distance using two robots in postprocessing. We further validated these algorithms with several ablation studies using a sperm whale visual encounter dataset collected by marine biologists.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

RoboticsAlgorithmsSperm WhaleVocalization, AnimalAnimalsAcousticsReinforcement, PsychologyDivingRadarComputer SimulationEquipment Design

Autor:innen und Institutionen

  • Project CETI, New York, NY, USA.
  • John A. Paulson School of Engineering and Applied Sciences, Harvard University, Cambridge, MA, USA.
  • Zoophysiology, Department of Biology, Aarhus University, 8000 Aarhus, Denmark.
  • College of Engineering, Carnegie Mellon University, Pittsburgh, PA, USA.
  • Department of Biology, Carleton University, Ottawa, Ontario, Canada.

ORCID-Profile: Ninad Jadhav, Sushmita Bhattacharya, Daniel Vogt, Yaniv Aluma, Pernille Tønnesen, Swarun Kumar, Shane Gero, Robert J Wood

Zitieren

APA

Jadhav, N., Bhattacharya, S., Vogt, D., Aluma, Y., Tønnesen, P., Prabhakara, A., Kumar, S., Gero, S., Wood, R. J., & Gil, S. (2024). Reinforcement learning–based framework for whale rendezvous via autonomous sensing robots. Science Robotics. https://doi.org/10.1126/scirobotics.adn7299

Vancouver

Jadhav N, Bhattacharya S, Vogt D, Aluma Y, Tønnesen P, Prabhakara A, et al. Reinforcement learning–based framework for whale rendezvous via autonomous sensing robots. Science Robotics. 2024. doi:10.1126/scirobotics.adn7299

BibTeX
@article{Jadhav2024,
  author = {Jadhav, Ninad and Bhattacharya, Sushmita and Vogt, Daniel and Aluma, Yaniv and Tønnesen, Pernille and Prabhakara, Akarsh and Kumar, Swarun and Gero, Shane and Wood, Robert J and Gil, Stephanie},
  title = {Reinforcement learning–based framework for whale rendezvous via autonomous sensing robots},
  journal = {Science Robotics},
  year = {2024},
  doi = {10.1126/scirobotics.adn7299}
}
Kostenlos starten →