Data-driven robotic sampling for marine ecosystem monitoring

Jnaneshwar Das, Frédéric Py, Julio B.J. Harvey, John P. Ryan, Alyssa Gellene, Rishi Graham, David A. Caron, Kanna Rajan, Gaurav S. Sukhatme

Research output: Contribution to journalArticle

34 Citations (Scopus)

Abstract

Robotic sampling is attractive in many field robotics applications that require persistent collection of physical samples for ex-situ analysis. Examples abound in the earth sciences in studies involving the collection of rock, soil, and water samples for laboratory analysis. In our test domain, marine ecosystem monitoring, detailed understanding of plankton ecology requires laboratory analysis of water samples, but predictions using physical and chemical properties measured in real-time by sensors aboard an autonomous underwater vehicle (AUV) can guide sample collection decisions. In this paper, we present a data-driven and opportunistic sampling strategy to minimize cumulative regret for batches of plankton samples acquired by an AUV over multiple surveys. Samples are labeled at the end of each survey, and used to update a probabilistic model that guides sampling during subsequent surveys. During a survey, the AUV makes irrevocable sample collection decisions online for a sequential stream of candidates, with no knowledge of the quality of future samples. In addition to extensive simulations using historical field data, we present results from a one-day field trial where beginning with a prior model learned from data collected and labeled in an earlier campaign, the AUV collected water samples with a high abundance of a pre-specified planktonic target. This is the first time such a field experiment has been carried out in its entirety in a data-driven fashion, in effect "closing the loop" on a significant and relevant ecosystem monitoring problem while allowing domain experts (marine ecologists) to specify the mission at a relatively high level.

Original languageEnglish (US)
Pages (from-to)1435-1452
Number of pages18
JournalInternational Journal of Robotics Research
Volume34
Issue number12
DOIs
StatePublished - Jan 1 2015
Externally publishedYes

Fingerprint

Aquatic ecosystems
Autonomous underwater vehicles
Ecosystem
Data-driven
Robotics
Monitoring
Sampling
Plankton
Underwater Vehicle
Water
Earth sciences
Ecology
Ecosystems
Chemical properties
Physical properties
Rocks
Soils
Sampling Strategy
Sensors
Field Experiment

Keywords

  • field robots
  • learning and adaptive systems
  • machine learning
  • marine robotics
  • Robotic sampling

ASJC Scopus subject areas

  • Software
  • Modeling and Simulation
  • Mechanical Engineering
  • Artificial Intelligence
  • Applied Mathematics
  • Electrical and Electronic Engineering

Cite this

Das, J., Py, F., Harvey, J. B. J., Ryan, J. P., Gellene, A., Graham, R., ... Sukhatme, G. S. (2015). Data-driven robotic sampling for marine ecosystem monitoring. International Journal of Robotics Research, 34(12), 1435-1452. https://doi.org/10.1177/0278364915587723

Data-driven robotic sampling for marine ecosystem monitoring. / Das, Jnaneshwar; Py, Frédéric; Harvey, Julio B.J.; Ryan, John P.; Gellene, Alyssa; Graham, Rishi; Caron, David A.; Rajan, Kanna; Sukhatme, Gaurav S.

In: International Journal of Robotics Research, Vol. 34, No. 12, 01.01.2015, p. 1435-1452.

Research output: Contribution to journalArticle

Das, J, Py, F, Harvey, JBJ, Ryan, JP, Gellene, A, Graham, R, Caron, DA, Rajan, K & Sukhatme, GS 2015, 'Data-driven robotic sampling for marine ecosystem monitoring', International Journal of Robotics Research, vol. 34, no. 12, pp. 1435-1452. https://doi.org/10.1177/0278364915587723
Das, Jnaneshwar ; Py, Frédéric ; Harvey, Julio B.J. ; Ryan, John P. ; Gellene, Alyssa ; Graham, Rishi ; Caron, David A. ; Rajan, Kanna ; Sukhatme, Gaurav S. / Data-driven robotic sampling for marine ecosystem monitoring. In: International Journal of Robotics Research. 2015 ; Vol. 34, No. 12. pp. 1435-1452.
@article{ab0a49b8d31d4451b196ef03764d8c5e,
title = "Data-driven robotic sampling for marine ecosystem monitoring",
abstract = "Robotic sampling is attractive in many field robotics applications that require persistent collection of physical samples for ex-situ analysis. Examples abound in the earth sciences in studies involving the collection of rock, soil, and water samples for laboratory analysis. In our test domain, marine ecosystem monitoring, detailed understanding of plankton ecology requires laboratory analysis of water samples, but predictions using physical and chemical properties measured in real-time by sensors aboard an autonomous underwater vehicle (AUV) can guide sample collection decisions. In this paper, we present a data-driven and opportunistic sampling strategy to minimize cumulative regret for batches of plankton samples acquired by an AUV over multiple surveys. Samples are labeled at the end of each survey, and used to update a probabilistic model that guides sampling during subsequent surveys. During a survey, the AUV makes irrevocable sample collection decisions online for a sequential stream of candidates, with no knowledge of the quality of future samples. In addition to extensive simulations using historical field data, we present results from a one-day field trial where beginning with a prior model learned from data collected and labeled in an earlier campaign, the AUV collected water samples with a high abundance of a pre-specified planktonic target. This is the first time such a field experiment has been carried out in its entirety in a data-driven fashion, in effect {"}closing the loop{"} on a significant and relevant ecosystem monitoring problem while allowing domain experts (marine ecologists) to specify the mission at a relatively high level.",
keywords = "field robots, learning and adaptive systems, machine learning, marine robotics, Robotic sampling",
author = "Jnaneshwar Das and Fr{\'e}d{\'e}ric Py and Harvey, {Julio B.J.} and Ryan, {John P.} and Alyssa Gellene and Rishi Graham and Caron, {David A.} and Kanna Rajan and Sukhatme, {Gaurav S.}",
year = "2015",
month = "1",
day = "1",
doi = "10.1177/0278364915587723",
language = "English (US)",
volume = "34",
pages = "1435--1452",
journal = "International Journal of Robotics Research",
issn = "0278-3649",
publisher = "SAGE Publications Inc.",
number = "12",

}

TY - JOUR

T1 - Data-driven robotic sampling for marine ecosystem monitoring

AU - Das, Jnaneshwar

AU - Py, Frédéric

AU - Harvey, Julio B.J.

AU - Ryan, John P.

AU - Gellene, Alyssa

AU - Graham, Rishi

AU - Caron, David A.

AU - Rajan, Kanna

AU - Sukhatme, Gaurav S.

PY - 2015/1/1

Y1 - 2015/1/1

N2 - Robotic sampling is attractive in many field robotics applications that require persistent collection of physical samples for ex-situ analysis. Examples abound in the earth sciences in studies involving the collection of rock, soil, and water samples for laboratory analysis. In our test domain, marine ecosystem monitoring, detailed understanding of plankton ecology requires laboratory analysis of water samples, but predictions using physical and chemical properties measured in real-time by sensors aboard an autonomous underwater vehicle (AUV) can guide sample collection decisions. In this paper, we present a data-driven and opportunistic sampling strategy to minimize cumulative regret for batches of plankton samples acquired by an AUV over multiple surveys. Samples are labeled at the end of each survey, and used to update a probabilistic model that guides sampling during subsequent surveys. During a survey, the AUV makes irrevocable sample collection decisions online for a sequential stream of candidates, with no knowledge of the quality of future samples. In addition to extensive simulations using historical field data, we present results from a one-day field trial where beginning with a prior model learned from data collected and labeled in an earlier campaign, the AUV collected water samples with a high abundance of a pre-specified planktonic target. This is the first time such a field experiment has been carried out in its entirety in a data-driven fashion, in effect "closing the loop" on a significant and relevant ecosystem monitoring problem while allowing domain experts (marine ecologists) to specify the mission at a relatively high level.

AB - Robotic sampling is attractive in many field robotics applications that require persistent collection of physical samples for ex-situ analysis. Examples abound in the earth sciences in studies involving the collection of rock, soil, and water samples for laboratory analysis. In our test domain, marine ecosystem monitoring, detailed understanding of plankton ecology requires laboratory analysis of water samples, but predictions using physical and chemical properties measured in real-time by sensors aboard an autonomous underwater vehicle (AUV) can guide sample collection decisions. In this paper, we present a data-driven and opportunistic sampling strategy to minimize cumulative regret for batches of plankton samples acquired by an AUV over multiple surveys. Samples are labeled at the end of each survey, and used to update a probabilistic model that guides sampling during subsequent surveys. During a survey, the AUV makes irrevocable sample collection decisions online for a sequential stream of candidates, with no knowledge of the quality of future samples. In addition to extensive simulations using historical field data, we present results from a one-day field trial where beginning with a prior model learned from data collected and labeled in an earlier campaign, the AUV collected water samples with a high abundance of a pre-specified planktonic target. This is the first time such a field experiment has been carried out in its entirety in a data-driven fashion, in effect "closing the loop" on a significant and relevant ecosystem monitoring problem while allowing domain experts (marine ecologists) to specify the mission at a relatively high level.

KW - field robots

KW - learning and adaptive systems

KW - machine learning

KW - marine robotics

KW - Robotic sampling

UR - http://www.scopus.com/inward/record.url?scp=84942603608&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84942603608&partnerID=8YFLogxK

U2 - 10.1177/0278364915587723

DO - 10.1177/0278364915587723

M3 - Article

AN - SCOPUS:84942603608

VL - 34

SP - 1435

EP - 1452

JO - International Journal of Robotics Research

JF - International Journal of Robotics Research

SN - 0278-3649

IS - 12

ER -