{"id":3984,"date":"2016-08-21T22:23:26","date_gmt":"2016-08-22T02:23:26","guid":{"rendered":"http:\/\/mseas.mit.edu\/?p=3984"},"modified":"2022-11-21T21:40:56","modified_gmt":"2022-11-22T02:40:56","slug":"path-planning-and-adaptive-sampling-in-the-coastal-ocean-2","status":"publish","type":"post","link":"https:\/\/mseas.mit.edu\/?p=3984","title":{"rendered":"Path Planning and Adaptive Sampling in the Coastal Ocean"},"content":{"rendered":"When humans or robots operate in complex dynamic environments, the planning of\npaths and the collection of observations are basic, indispensable problems. In the\noceanic and atmospheric environments, the concurrent use of multiple mobile sensing\nplatforms in unmanned missions is growing very rapidly. Opportunities for a\nparadigm shift in the science of autonomy involve the development of fundamental\ntheories to optimally collect information, learn, collaborate and make decisions under\nuncertainty while persistently adapting to and utilizing the dynamic environment.\nTo address such pressing needs, this thesis derives governing equations and develops\nrigorous methodologies for optimal path planning and optimal sampling using collaborative\nswarms of autonomous mobile platforms. The application focus is the coastal\nocean where currents can be much larger than platform speeds, but the fundamental\nresults also apply to other dynamic environments.\n\nWe first undertake a theoretical synthesis of minimum-time control of vehicles operating\nin general dynamic flows. Using various ideas rooted in non-smooth calculus,\nwe prove that an unsteady Hamilton-Jacobi equation governs the forward reachable\nsets in any type of Lipschitz-continuous flow. Next, we show that with a suitable\nmodification to the Hamiltonian, the results can be rigorously generalized to perform\ntime-optimal path planning with anisotropic motion constraints and with moving obstacles\nand unsafe \u2018forbidden\u2019 regions. We then derive a level-set methodology for\ndistance-based coordination of swarms of vehicles operating in minimum time within\nstrong and dynamic ocean currents. The results are illustrated for varied fluid and\nocean flow simulations. Finally, the new path planning system is applied to swarms\nof vehicles operating in the complex geometry of the Philippine Archipelago, utilizing\nrealistic multi-scale current predictions from a data-assimilative ocean modeling\nsystem.\n\nIn the second part of the thesis, we derive a theory for adaptive sampling that exploits\nthe governing nonlinear dynamics of the system and captures the non-Gaussian\nstructure of the random state fields. Optimal observation locations are determined\nby maximizing the mutual information between the candidate observations and the\nvariables of interest. We develop a novel Bayesian smoother for high-dimensional continuous stochastic fields governed by general nonlinear dynamics. This smoother\ncombines the adaptive reduced-order Dynamically-Orthogonal equations with Gaussian\nMixture Models, extending linearized Gaussian backward pass updates to a nonlinear,\nnon-Gaussian setting. The Bayesian information transfer, both forward and\nbackward in time, is efficiently carried out in the evolving dominant stochastic subspace.\nBuilding on the foundations of the smoother, we then derive an efficient\ntechnique to quantify the spatially and temporally varying mutual information field\nin general nonlinear dynamical systems. The globally optimal sequence of future sampling\nlocations is rigorously determined by a novel dynamic programming approach\nthat combines this computation of mutual information fields with the predictions of\nthe forward reachable set. All the results are exemplified and their performance is\nquantitatively assessed using a variety of simulated fluid and ocean flows.\n\nThe above novel theories and schemes are integrated so as to provide real-time\ncomputational intelligence for collaborative swarms of autonomous sensing vehicles.\nThe integrated system guides groups of vehicles along predicted optimal trajectories\nand continuously improves field estimates as the observations predicted to be most\ninformative are collected and assimilated. The optimal sampling locations and optimal\ntrajectories are continuously forecast, all in an autonomous and coordinated\nfashion.","protected":false},"excerpt":{"rendered":"<p>When humans or robots operate in complex dynamic environments, the planning of paths and the collection of observations are basic, indispensable problems. In the oceanic and atmospheric environments, the concurrent use of multiple mobile sensing platforms in unmanned missions is growing very rapidly. Opportunities for a paradigm shift in the science of autonomy involve the [&hellip;]<\/p>\n","protected":false},"author":2,"featured_media":0,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[42,5,45],"tags":[219,223,241],"class_list":["post-3984","post","type-post","status-publish","format-standard","hentry","category-meche-theses","category-publications","category-ph-d-theses","tag-a-mission","tag-learns","tag-phile"],"_links":{"self":[{"href":"https:\/\/mseas.mit.edu\/index.php?rest_route=\/wp\/v2\/posts\/3984","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/mseas.mit.edu\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/mseas.mit.edu\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/mseas.mit.edu\/index.php?rest_route=\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/mseas.mit.edu\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=3984"}],"version-history":[{"count":3,"href":"https:\/\/mseas.mit.edu\/index.php?rest_route=\/wp\/v2\/posts\/3984\/revisions"}],"predecessor-version":[{"id":6266,"href":"https:\/\/mseas.mit.edu\/index.php?rest_route=\/wp\/v2\/posts\/3984\/revisions\/6266"}],"wp:attachment":[{"href":"https:\/\/mseas.mit.edu\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=3984"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/mseas.mit.edu\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=3984"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/mseas.mit.edu\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=3984"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}