Learning excursion sets of vector-valued Gaussian random fields for autonomous ocean sampling. (English) Zbl 1478.62354

Summary: 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 provides 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.


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


[1] Azzimonti, D., Bect, J., Chevalier, C. and Ginsbourger, D. (2016). Quantifying uncertainties on excursion sets under a Gaussian random field prior. SIAM/ASA J. Uncertain. Quantificat. 4 850-874. · Zbl 1352.62144 · doi:10.1137/141000749
[2] Azzimonti, D., Ginsbourger, D., Chevalier, C., Bect, J. and Richet, Y. (2021). Adaptive design of experiments for Conservative estimation of excursion sets. Technometrics 63 13-26. · doi:10.1080/00401706.2019.1693427
[3] Bect, J., Bachoc, F. and Ginsbourger, D. (2019). A supermartingale approach to Gaussian process based sequential design of experiments. Bernoulli 25 2883-2919. · Zbl 1428.62369 · doi:10.3150/18-BEJ1074
[4] Bect, J., Ginsbourger, D., Li, L., Picheny, V. and Vazquez, E. (2012). Sequential design of computer experiments for the estimation of a probability of failure. Stat. Comput. 22 773-793. · Zbl 1252.62081 · doi:10.1007/s11222-011-9241-4
[5] Bellingham, J. G. and Rajan, K. (2007). Robotics in remote and hostile environments. Science 318 1098-1102.
[6] Bhattacharjya, D., Eidsvik, J. and Mukerji, T. (2013). The value of information in portfolio problems with dependent projects. Decis. Anal. 10 341-351. · Zbl 1398.91505 · doi:10.1287/deca.2013.0277
[7] Binney, J., Krause, A. and Sukhatme, G. S. (2013). Optimizing waypoints for monitoring spatiotemporal phenomena. The International Journal of Robotics Research 32 873-888.
[8] Bolin, D. and Lindgren, F. (2015). Excursion and contour uncertainty regions for latent Gaussian models. J. R. Stat. Soc. Ser. B. Stat. Methodol. 77 85-106. · Zbl 1414.62332 · doi:10.1111/rssb.12055
[9] Bueso, M., Angulo, J. and Alonso, F. (1998). A state-space model approach to optimum spatial sampling design based on entropy. Environ. Ecol. Stat. 5 29-44.
[10] Chevalier, C., Emery, X. and Ginsbourger, D. (2015). Fast update of conditional simulation ensembles. Math. Geosci. 47 771-789. · Zbl 1323.86020
[11] Chevalier, C., Ginsbourger, D. and Emery, X. (2013). Corrected kriging update formulae for batch-sequential data assimilation. In Mathematics of Planet Earth (E. Pardo-Igúzquiza, C. Guardiola-Albert, J. Heredia, L. Moreno-Merino, J. Durán and J. Vargas-Guzmán, eds.). Lecture Notes in Earth System Sciences. Springer, Berlin.
[12] Chevalier, C., Ginsbourger, D., Bect, J. and Molchanov, I. (2013). Estimating and quantifying uncertainties on level sets using the Vorob’ev expectation and deviation with Gaussian process models. In MODa 10 Advances in Model-Oriented Design and Analysis (D. Uciński, A. C. Atkinson and C. Patan, eds.) Physica-Verlag, Heidelberg.
[13] Chevalier, C., Ginsbourger, D., Bect, J., Vazquez, E., Picheny, V. and Richet, Y. (2014). Fast parallel kriging-based stepwise uncertainty reduction with application to the identification of an excursion set. Technometrics 56 455-465. · doi:10.1080/00401706.2013.860918
[14] Costa, M. J., Pinto, J., Dias, P. S., Pereira, J., Lima, K., Ribeiro, M., Sousa, J. B., Lukaczyk, T., Mendes, R. et al. (2018). Field report: Exploring fronts with multiple robots. In 2018 IEEE/OES Autonomous Underwater Vehicle Symposium.
[15] Das, J., Py, F., Maughan, T., Messie, M., O’Reilly, T., Ryan, J., Sukhatme, G. S. and Rajan, K. (2012). Coordinated sampling of dynamic oceanographic features with AUVs and drifters. International Journal of Robotics Research 31 626-646.
[16] Das, J., Py, F., Harvey, J. B. J., Ryan, J. P., Gellene, A., Graham, R., Caron, D. A., Rajan, K. and Sukhatme, G. S. (2015). Data-driven robotic sampling for marine ecosystem monitoring. The International Journal of Robotics Research 34 1435-1452.
[17] Davidson, F., Alvera-Azcárate, A., Barth, A., Brassington, G. B., Chassignet, E. P., Clementi, E., De Mey-Frémaux, P., Divakaran, P., Harris, C. et al. (2019). Synergies in operational oceanography: The intrinsic need for sustained ocean observations. Front. Mar. Sci. 6 450.
[18] Eidsvik, J., Mukerji, T. and Bhattacharjya, D. (2015). Value of Information in the Earth Sciences: Integrating Spatial Modeling and Decision Analysis. Cambridge Univ. Press, Cambridge.
[19] Ferreira, A. S., Costa, M., Py, F., Pinto, J., Silva, M. A., Nimmo-Smith, A., Johansen, T. A., de Sousa, J. B. and Rajan, K. (2019). Advancing multi-vehicle deployments in oceanographic field experiments. Autonomous Robots 43 1555-1574.
[20] Fossum, T. O., Eidsvik, J., Ellingsen, I., Alver, M. O., Fragoso, G. M., Johnsen, G., Mendes, R., Ludvigsen, M. and Rajan, K. (2018). Information-driven robotic sampling in the coastal ocean. J. Field Robot. 35 1101-1121.
[21] Fossum, T. O., Fragoso, G. M., Davies, E. J., Ullgren, J. E., Mendes, R., Johnsen, G., Ellingsen, I., Eidsvik, J., Ludvigsen, M. et al. (2019). Toward adaptive robotic sampling of phytoplankton in the coastal ocean. Sci Robot 4. · doi:10.1126/scirobotics.aav3041
[22] Fossum, T. O., Travelletti, C., Eidsvik, J., Ginsbourger, D. and Rajan, K. (2021). Supplement to “Learning excursion sets of vector-valued Gaussian random fields for autonomous ocean sampling.” https://doi.org/10.1214/21-AOAS1451SUPPA
[23] Fossum, T. O., Travelletti, C., Eidsvik, J., Ginsbourger, D. and Rajan, K. (2021). Source code for “Learning excursion sets of vector-valued Gaussian random fields for autonomous ocean sampling”—The MESLAS package. https://doi.org/10.1214/21-AOAS1451SUPPB
[24] French, J. P. and Sain, S. R. (2013). Spatio-temporal exceedance locations and confidence regions. Ann. Appl. Stat. 7 1421-1449. · Zbl 1283.62226 · doi:10.1214/13-AOAS631
[25] Genton, M. G. and Kleiber, W. (2015). Cross-covariance functions for multivariate geostatistics. Statist. Sci. 30 147-163. · Zbl 1332.86010 · doi:10.1214/14-STS487
[26] Genz, A. and Bretz, F. (2009). Computation of Multivariate Normal and t Probabilities. Lecture Notes in Statistics 195. Springer, Dordrecht. · Zbl 1204.62088 · doi:10.1007/978-3-642-01689-9
[27] Ginsbourger, D. (2018). Wiley StatsRef: Statistics Reference Online. Sequential Design of Computer Experiments 1-9.
[28] Gneiting, T., Kleiber, W. and Schlather, M. (2010). Matérn cross-covariance functions for multivariate random fields. J. Amer. Statist. Assoc. 105 1167-1177. · Zbl 1390.62194 · doi:10.1198/jasa.2010.tm09420
[29] Gottlieb, J., Graham, R., Maughan, T., Py, F., Elkaim, G. and Rajan, K. (2012). An experimental momentum-based front detection for autonomous underwater vehicles. In IEEE International Conference on Robotics and Automation (ICRA).
[30] Hopkins, J., Lucas, M., Dufau, C., Sutton, M., Stum, J., Lauret, O. and Channelliere, C. (2013). Detection and variability of the Congo river plume from satellite derived sea surface temperature, salinity, ocean colour and sea level. Remote Sens. Environ. 139 365-385.
[31] Karaman, S. and Frazzoli, E. (2011). Sampling-based algorithms for optimal motion planning. The International Journal of Robotics Research 30 846-894. · Zbl 1220.91006
[32] Krige, D. G. (1951). A statistical approach to some basic mine valuation problems on the witwatersrand. J. of the Chem., Metal. and Mining Soc. of South Africa 52 119-139.
[33] Le Gratiet, L. and Cannamela, C. (2015). Cokriging-based sequential design strategies using fast cross-validation techniques for multi-fidelity computer codes. Technometrics 57 418-427. · doi:10.1080/00401706.2014.928233
[34] Lermusiaux, P. F. J. (2006). Uncertainty estimation and prediction for interdisciplinary ocean dynamics. J. Comput. Phys. 217 176-199. · Zbl 1146.86002 · doi:10.1016/j.jcp.2006.02.010
[35] Low, K. H., Dolan, J. M. and Khosla, P. K. (2009). Information-theoretic approach to efficient adaptive path planning for mobile robotic environmental sensing. Proceedings of the International Conference on Automated Planning and Scheduling 233-240.
[36] Mellucci, C., Menon, P. P., Edwards, C. and Challenor, P. (2018). Oceanic feature boundary mapping with an autonomous underwater glider. In 2018 Annual American Control Conference (ACC) 5338-5343. IEEE, New York.
[37] Molchanov, I. (2005). Theory of Random Sets. Probability and Its Applications (New York). Springer London, Ltd., London. · Zbl 1109.60001
[38] Müller, W. G. (2007). Collecting Spatial Data: Optimum Design of Experiments for Random Fields, third revised and extended ed. Contributions to Statistics. Physica-Verlag, Heidelberg.
[39] Pinto, J., Mendes, R., da Silva, J. C. B., Dias, J. M. and de Sousa, J. B. (2018). Multiple autonomous vehicles applied to plume detection and tracking. In 2018 OCEANS—MTS/IEEE Kobe Techno-Oceans (OTO) 1-6.
[40] Poloczek, M. U., Wang, J. and Frazier, P. I. (2017). Multi-information source optimization. In Advances in Neural Information Processing Systems 30.
[41] Powell, W. B. (2016). Perspectives of approximate dynamic programming. Ann. Oper. Res. 241 319-356. · Zbl 1348.90612 · doi:10.1007/s10479-012-1077-6
[42] Py, F., Rajan, K. and McGann, C. (2010). A systematic agent framework for situated autonomous systems. In 9th International Conf. on Autonomous Agents and Multiagent Systems (AAMAS).
[43] Rajan, K. and Py, F. (2012). T-REX: Partitioned inference for AUV mission control. In Further Advances in Unmanned Marine Vehicles (G. N. Roberts and R. Sutton, eds.) The Institution of Engineering and Technology (IET).
[44] Rajan, K., Py, F. and Berreiro, J. (2012). Towards deliberative control in marine robotics. In Marine Robot Autonomy (M. Seto, ed.) Springer, Berlin.
[45] Richardson, R. A. (2017). Sparsity in nonlinear dynamic spatiotemporal models using implied advection. Environmetrics 28 e2456. · doi:10.1002/env.2456
[46] Ryan, J. P., Kudela, R. M., Birch, J. M., Blum, M., Bowers, H. A., Chavez, F. P., Doucette, G. J., Hayashi, K., Marin, R. III et al. (2017). Causality of an extreme harmful algal bloom in Monterey Bay, California, during the 2014-2016 northeast Pacific warm anomaly. Geophys. Res. Lett. 44 5571-5579.
[47] Sahu, S. K. and Challenor, P. (2008). A space-time model for joint modeling of ocean temperature and salinity levels as measured by Argo floats. Environmetrics 19 509-528. · doi:10.1002/env.895
[48] Santner, T. J., Williams, B. J. and Notz, W. I. (2003). The Design and Analysis of Computer Experiments. Springer Series in Statistics. Springer, New York. · Zbl 1041.62068 · doi:10.1007/978-1-4757-3799-8
[49] Sigrist, F., Künsch, H. R. and Stahel, W. A. (2015). Stochastic partial differential equation based modelling of large space-time data sets. J. R. Stat. Soc. Ser. B. Stat. Methodol. 77 3-33. · Zbl 1414.62405 · doi:10.1111/rssb.12061
[50] Singh, A., Krause, A., Guestrin, C. and Kaiser, W. J. (2009). Efficient informative sensing using multiple robots. J. Artificial Intelligence Res. 34 707-755. · Zbl 1192.68693 · doi:10.1613/jair.2674
[51] Smith, R. N., Py, F., Cooksey, P., Sukhatme, G. and Rajan, K. (2014). Adaptive path planning for tracking ocean fronts with an autonomous underwater vehicle. In Intnl. Symp. on Experimental Robotics (ISER).
[52] Sousa, A., Madureira, L., Coelho, J., Pinto, J., Pereira, J., Sousa, J. and Dias, P. (2012). LAUV: The man-portable autonomous underwater vehicle. In Navigation, Guidance and Control of Underwater Vehicles 3 268-274.
[53] Stein, M. L. (1999). Interpolation of Spatial Data: Some Theory for Kriging. Springer Series in Statistics. Springer, New York. · Zbl 0924.62100 · doi:10.1007/978-1-4612-1494-6
[54] Stroh, R. (2018). Planification d’expériences numériques en multi-fidélité: Application à un simulateur d’incendies. Ph.D. thesis, Univ. Paris-Saclay.
[55] Vargas-Guzmán, J. A. and Jim Yeh, T. C. (1999). Sequential kriging and cokriging: Two powerful geostatistical approaches. Stochastic Environmental Research and Risk Assessment 13 416-435. · Zbl 0956.62063
[56] Wackernagel, H. (2003). Multivariate Geostatistics: An Introduction with Applications. Springer, Berlin. · Zbl 1015.62128
[57] Wikle, C. K., Milliff, R. F., Herbei, R. and Leeds, W. B. (2013). Modern statistical methods in oceanography: A hierarchical perspective. Statist. Sci. 28 466-486. · Zbl 1331.86031 · doi:10.1214/13-STS436
[58] Zidek, J. V. and Zimmerman, D. L. (2019). Monitoring network design. In Handbook of Environmental and Ecological Statistics. Chapman & Hall/CRC Handb. Mod. Stat. Methods 499-522. CRC Press, Boca Raton, FL
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.