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.
AndersonCRSapianoMRPPrasad. (2010) Predicting potentially toxigenic Pseudo-nitzschia blooms in the Chesapeake Bay. Journal of Marine Systems83(3): 12–140.
2.
AnthonyDElbaumSLorenzA. (2014) On crop height estimation with UAVs. In: IEEE/RSJ international conference on intelligent robots and systems (IROS), Chicago, USA, 14–18 September, pp. 4805–4812. IEEE.
3.
AudibertJYMunosRSzepesváriC (2009) Exploration–exploitation tradeoff using variance estimates in multi-armed bandits. Theoretical Computer Science410(19): 1876–1902.
4.
AzimiJJalaliAZhang-FernX (2012) Hybrid batch Bayesian optimization. In: Proceedings of the 29th international conference on machine learning (ICML), Edinburgh, UK, 26 June–1 July, pp. 1215–1222. Madison: Omnipress.
5.
BachF (2011) Learning with submodular functions: A convex optimization perspective. Report, arXiv: abs/1111.6453.
6.
BatalinMARahimiMYuY. (2004) Call and response: Experiments in sampling the environment. In: Proceedings of the 2nd international conference on embedded networked sensor systems, Baltimore, USA, 3–5 November, pp. 25–38. New York: ACM.
7.
BateniMHajiaghayiMZadimoghaddamM (2010) Submodular secretary problem and extensions. In: Proceedings of the approximation, randomization, and combinatorial optimization. Algorithms and techniques: 13th international workshop, APPROX 2010, and 14th international workshop, RANDOM 2010, Barcelona, Spain, 1–3 September 2010, vol. 6302, pp. 39–52. Berlin: Springer Science & Business Media.
8.
BinneyJKrauseASukhatmeGS (2013) Optimizing waypoints for monitoring spatiotemporal phenomena. International Journal of Robotics Research32(8): 873–888.
9.
BrysonMReidASRamosFT. (2010) Airborne vision-based mapping and classification of large farmland environments. Journal of Field Robotics27(5): 632–655.
10.
ChenYKrauseA (2013) Near-optimal batch mode active learning and adaptive submodular optimization. In: International conference on machine learning (ICML), vol. 28, Atlanta, USA, 16–21 June, pp. 160–168.
11.
DasJPyFMaughanT. (2012) Coordinated sampling of dynamic oceanographic features with underwater vehicles and drifters. International Journal of Robotics Research31(5): 626–646.
12.
DasJRajanKFrolovS. (2010) Towards marine bloom trajectory prediction for AUV mission planning. In: IEEE International Conference on Robotics and Automation, Anchorage, USA, 3–7 May, pp. 4784–4790. IEEE.
13.
DesautelsTKrauseABurdickJ (2014) Parallelizing exploration–exploitation tradeoffs with gaussian process bandit optimization. The Journal of Machine Learning Research15(1): 3873–3923.
14.
ElithJPhillipsSJHastieT. (2011) A statistical explanation of MaxEnt for ecologists. Diversity and Distributions17(1): 43–57.
15.
FergusonT (1989) Who solved the secretary problem?Statistical Science4: 282–289.
16.
FoxMLongDPyF. (2007) In situ analysis for intelligent control. In: Proceedings of IEEE/OES OCEANS conference, Aberdeen, UK, 18–21 June, pp. 1–6. IEEE.
17.
Garcia-OlayaAPyFDasJ. (2012) An on-line utility based multi-criteria approach for sampling dynamic ocean fields. IEEE Journal of Oceanic Engineering37(2): 185–203.
18.
GargSSinghARamosF (2012) Learning non-stationary space-time models for environmental monitoring. In: 26th conference on artificial intelligence (AAAI), special track on computational sustainability and AI, Ontario, Canada, 22–26 July. New York: ACM.
19.
GirdharYDudekG (2009) Optimal online data sampling or how to hire the best secretaries. In: Canadian conference on computer and robot vision, Kelowna, Canada, 25–27 May, pp. 292–298. IEEE.
20.
HarveyJBJ (2014) A 96-well plate format for detection of marine zooplankton with the sandwich hybridization assay. In: StrickerSCarrollD (eds) Methods in Molecular Biology: Developmental Biology of the Sea Urchin and Other Marine Invertebrates. New York: Humana Press Incorporation.
21.
HarveyJBRyanJPMarinR. (2012a) Robotic sampling, in situ monitoring and molecular detection of marine zooplankton. Journal of Experimental Marine Biology and Ecology413: 60–70.
22.
HarveyJBJZhangYRyanJP (2012b) AUVs for ecological studies of marine plankton communities: Intelligent algorithms on Dorado and Tethys AUVs enable precise water sampling for plankton research. Sea Technology, September.
23.
HitzGPomerleauFGarneauME. (2012) Autonomous inland water monitoring: design and application of a surface vessel. Robotics & Automation Magazine, IEEE19(1): 62–72.
24.
JulianBJAngermannMSchwagerM. (2012) Distributed robotic sensor networks: An information theoretic approach. International Journal of Robotics Research31(10): 1134–1154.
25.
KimYShellDAHoC. (2013) Spatial interpolation for robotic sampling: Uncertainty with two models of variance. In: DesaiJPDudekGKhatibO. (eds) Experimental Robotics: Springer Tracts in Advanced Robotics, Volume 88. Springer, pp. 759–774.
26.
KrauseAGuestrinC (2007) Nonmyopic active learning of Gaussian processes: An exploration–exploitation approach. In: Proceedings of the 24th international conference on machine learning (ICML), Corvalis, USA, 20–24 June, pp. 449–456. New York: ACM.
27.
KrauseAGuestrinCGuptaA. (2006) Near-optimal sensor placements: Maximizing information while minimizing communication cost. In: Proceedings of the 5th international conference on information processing in sensor networks, Nashville, USA, 19–21 April, pp. 2–10. New York: ACM.
28.
LowKHDolanJKhoslaP (2009) Information-theoretic approach to efficient adaptive path planning for mobile robotic environmental sensing. In: Proceedings of the 19th international conference on automated planning and scheduling (ICAPS-09), Thessaloniki, Greece, 19–23 September. Menlo Park: AAAI.
29.
MATLAB (2013) version 8.1.0.604, R2013a. Natick, MA: The MathWorks Incorporation.
30.
RasmussenCE (2006) Gaussian Processes for Machine Learning. Cambridge, MA: MIT Press.
ReidARamosFSukkariehS (2013) Bayesian Fusion for Multi-Modal Aerial Images. In: Proceedings of robotics: science and systems, Berlin, Germany, 24–28 June, pp. 1–8. Cambridge, MA: MIT Press.
33.
RyanJHarveyJZhangY. (2014a) Distributions of invertebrate larvae and phytoplankton in a coastal upwelling system retention zone and peripheral front. Journal of Experimental Marine Biology and Ecology459(0): 51–60.
34.
RyanJMcManusMKudelaR. (2014b) Boundary influences on HAB phytoplankton ecology in a stratification-enhanced upwelling shadow. Deep Sea Research Part II: Topical Studies in Oceanography101(0): 63–79.
35.
RyanJPJohnsonSShermanA. (2010) Intermediate nepheloid layers as conduits of larval transport. Limnology & Oceanograhy Methods8: 394–402.
36.
ShkurtiFXuAMeghjaniM. (2012) Multi-domain monitoring of marine environments using a heterogeneous robot team. In: IEEE/RSJ international conference on intelligent robots and systems (IROS), pp. 1747–1753.
37.
SinghAKrauseAGuestrinC. (2009) Efficient informative sensing using multiple robots. Journal of Artificial Intelligence Research34: 707–755.
38.
SinghARamosFDurrant-WhyteH. (2010) Modeling and decision making in spatio-temporal processes for environmental surveillance. In: IEEE international conference on robotics and automation (ICRA), Anchorage, USA, 3–7 May, pp. 5490–5497. IEEE.
39.
SrinivasNKrauseAKakadeS. (2012) Information-theoretic regret bounds for Gaussian process optimization in the bandit setting. IEEE Transactions on Information Theory58(5): 3250–3265.
40.
SrinivasNKrauseAKakadeSM. (2009) Gaussian process optimization in the bandit setting: no regret and experimental design. In: Proceedings of the 27th international conference on machine learning (ICML), Haifa, Israel, 21–24 June, pp. 1015–1022. Madison: Omnipress.
41.
StealeyMJSinghABatalinMA. (2008) NIMS-AQ: A novel system for autonomous sensing of aquatic environments. In: IEEE international conference on robotics and automation (ICRA), pp. 621–628.
42.
TokekarPBhadauriaDStudenskiA. (2010) A robotic system for monitoring carp in Minnesota lakes. Journal of Field Robotics27(6): 779–789.
43.
TokekarPVander HookJMullaD. (2013) Sensor planning for a symbiotic UAV and UGV system for precision agriculture. In: IEEE/RSJ international conference on intelligent robots and systems (IROS), Tokyo, Japan, 3–7 November, pp. 5321–5326. IEEE.
44.
ZhangBSukhatmeGS (2007) Adaptive sampling for estimating a scalar field using a robotic boat and a sensor network. In: IEEE international conference on robotics and automation (ICRA), Rome, Italy, 10–14 April. IEEE.
45.
ZhangYMcEwenRSRyanJPBellinghamJG (2010) Design and tests of an adaptive triggering method for capturing peak samples in a thin phytoplankton layer by an autonomous underwater vehicle. IEEE Journal of Oceanic Engineering35: 785–796.
46.
ZhangYRyanJPBellinghamJG. (2012) Autonomous detection and sampling of water types and fronts in a coastal upwelling system by an autonomous underwater vehicle. Limnology Oceanography: Methods10: 934–951.
