Learning excursion sets of vector-valued Gaussian random fields for autonomous ocean sampling

DOI10.1214/21-AOAS1451zbMATH Open1478.62354arXiv2007.03722OpenAlexW3178168838MaRDI QIDQ2245134FDOQ2245134

Authors: Trygve Olav Fossum, Cédric Travelletti, Jo Eidsvik, D. Ginsbourger, Kanna Rajan 

Publication date: 15 November 2021

Published in: The Annals of Applied Statistics (Search for Journal in Brave)

Abstract: Improving and optimizing oceanographic sampling is a crucial task for marine science and maritime resource management. Faced with limited resources in understanding processes in the water-column, the combination of statistics and autonomous systems provide new opportunities for experimental design. In this work we develop efficient spatial sampling methods for characterizing regions defined by simultaneous exceedances above prescribed thresholds of several responses, with an application focus on mapping coastal ocean phenomena based on temperature and salinity measurements. Specifically, we define a design criterion based on uncertainty in the excursions of vector-valued Gaussian random fields, and derive tractable expressions for the expected integrated Bernoulli variance reduction in such a framework. We demonstrate how this criterion can be used to prioritize sampling efforts at locations that are ambiguous, making exploration more effective. We use simulations to study and compare properties of the considered approaches, followed by results from field deployments with an autonomous underwater vehicle as part of a study mapping the boundary of a river plume. The results demonstrate the potential of combining statistical methods and robotic platforms to effectively inform and execute data-driven environmental sampling.

Full work available at URL: https://arxiv.org/abs/2007.03722

Recommendations

• Adaptive on-line planning of environmental sampling missions with a team of cooperating autonomous underwater vehicles
• Cooperative adaptive sampling of random fields with partially known covariance
• Mobile Robot Sensing for Environmental Applications
• Incorporating prior knowledge in observability-based path planning for ocean sampling
• Adaptive modeling, adaptive data assimilation and adaptive sampling

zbMATH Keywords

experimental designGaussian processesexcursion setsocean samplingautonomous robotsadaptive information gathering

Mathematics Subject Classification ID

Gaussian processes (60G15) Sampling theory, sample surveys (62D05) Optimal statistical designs (62K05) Applications of statistics to environmental and related topics (62P12) Hydrology, hydrography, oceanography (86A05)

Cites Work

• Computation of multivariate normal and \(t\) probabilities
• Interpolation of spatial data. Some theory for kriging
• The design and analysis of computer experiments.
• Matérn cross-covariance functions for multivariate random fields
• Spatio-temporal exceedance locations and confidence regions
• Excursion and contour uncertainty regions for latent Gaussian models
• Theory of Random Sets
• Sequential design of computer experiments for the estimation of a probability of failure
• Quantifying uncertainties on excursion sets under a Gaussian random field prior
• A supermartingale approach to Gaussian process based sequential design of experiments
• Fast update of conditional simulation ensembles
• Cross-covariance functions for multivariate geostatistics
• Title not available (Why is that?)
• Uncertainty estimation and prediction for interdisciplinary ocean dynamics
• Perspectives of approximate dynamic programming
• The value of information in portfolio problems with dependent projects
• Sequential kriging and cokriging: Two powerful geostatistical approaches
• Incremental sampling-based algorithms for a class of pursuit-evasion games
• Stochastic partial differential equation based modelling of large space-time data sets
• Efficient informative sensing using multiple robots
• Modern statistical methods in oceanography: a hierarchical perspective

Cited In (4)

• Uncertainty Quantification and Experimental Design for Large-Scale Linear Inverse Problems under Gaussian Process Priors
• Using an autonomous underwater vehicle with onboard stochastic advection-diffusion models to map excursion sets of environmental variables
• Adaptive spatial designs minimizing the integrated Bernoulli variance in spatial logistic regression models -- with an application to benthic habitat mapping
• Model-robust Bayesian design through generalised additive models for monitoring submerged shoals

This page was built for publication: Learning excursion sets of vector-valued Gaussian random fields for autonomous ocean sampling

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2245134)