ALBERT

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Mechanical and Oceanographic Engineering at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution February 2023.

Description: To predict and mitigate anthropogenic impacts on the ocean, we must understand the underlying systems that govern the ocean’s response to inputs (e.g. carbon dioxide, pollutants). Analytical models can be used to generate predictions and simulate intervention strategies, but they must be grounded with empirical observations. Unfortunately, there exists a technological gap: in situ instrumentation is often lacking or nonexistent for key parameters influenced by anthropogenic inputs. While discrete bottle samples can be collected and analyzed for these parameters, their limited spatiotemporal resolution constrains scientific inquiry. To help fill the technological gap, this dissertation presents the development of instrumentation for the ocean inorganic carbon system and microplastics. The first few chapters present the development process of CSPEC, a deep-sea laser spectrometer designed to measure the ocean carbon system through alternating measurements of the partial pressure of carbon dioxide (pCO2) and dissolved inorganic carbon (DIC). CSPEC uses tunable diode laser absorption spectroscopy (TDLAS) to measure the CO2 content of dissolved gas extracted via a membrane inlet. Chapter 2 derives membrane equilibration dynamics from first principles, thus enabling informed design decisions. The analytical results showed that cross-sensitivity to other dissolved gases can be introduced by the equilibration method, regardless of the specificity of the gas-side instrumentation. A new method, hybrid equilibration, leverages the membrane equilibration dynamics to improve time response without incurring cross-sensitivity. Chapter 3 presents POCO, a surface pCO2 instrument that employs TDLAS and a depth-compatible membrane inlet. Through laboratory and field-testing, POCO demonstrated that hybrid equilibration overcame the gas flux limitation of deep-sea membrane inlets. Chapter 4 presents CSPEC, which successfully mapped the carbon system near different hydrothermal features at 2000 m in Guaymas Basin, becoming one of the first DIC instruments field-tested at depth. Chapter 5 introduces impedance spectroscopy for quantifying microplastics directly in water. Microplastics were successfully counted, sized, and differentiated from biology in the laboratory: a step toward in situ quantification. The analytical tools and measurement systems presented in this dissertation represent a significant step towards increasing the spatiotemporal resolution of carbon system and microplastic measurements, thus enabling broader scientific inquiry in the future.

Description: This research was supported by the following funding sources: NSF Grant # OCE-1454067 NSF Grant # OCE-184-2053 Link Foundation Ocean Engineering and Instrumentation Ph.D. Fellowship MITMartin Family Society of Fellows for Sustainability Richard Saltonstall Charitable Foundation National Academies Keck Future Initiative (NAFKI DBS13)

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution September 2022.

Description: How do we collect observational data that reveal fundamental properties of scientific phenomena? This is a key challenge in modern scientific discovery. Scientific phenomena are complex—they have high-dimensional and continuous state, exhibit chaotic dynamics, and generate noisy sensor observations. Additionally, scientific experimentation often requires significant time, money, and human effort. In the face of these challenges, we propose to leverage autonomous decision-making to augment and accelerate human scientific discovery. Autonomous decision-making in scientific domains faces an important and classical challenge: balancing exploration and exploitation when making decisions under uncertainty. This thesis argues that efficient decision-making in real-world, scientific domains requires task-targeted exploration—exploration strategies that are tuned to a specific task. By quantifying the change in task performance due to exploratory actions, we enable decision-makers that can contend with highly uncertain real-world environments, performing exploration parsimoniously to improve task performance. The thesis presents three novel paradigms for task-targeted exploration that are motivated by and applied to real-world scientific problems. We first consider exploration in partially observable Markov decision processes (POMDPs) and present two novel planners that leverage task-driven information measures to balance exploration and exploitation. These planners drive robots in simulation and oceanographic field trials to robustly identify plume sources and track targets with stochastic dynamics. We next consider the exploration- exploitation trade-off in online learning paradigms, a robust alternative to POMDPs when the environment is adversarial or difficult to model. We present novel online learning algorithms that balance exploitative and exploratory plays optimally under real-world constraints, including delayed feedback, partial predictability, and short regret horizons. We use these algorithms to perform model selection for subseasonal temperature and precipitation forecasting, achieving state-of-the-art forecasting accuracy. The human scientific endeavor is poised to benefit from our emerging capacity to integrate observational data into the process of model development and validation. Realizing the full potential of these data requires autonomous decision-makers that can contend with the inherent uncertainty of real-world scientific domains. This thesis highlights the critical role that task-targeted exploration plays in efficient scientific decision-making and proposes three novel methods to achieve task-targeted exploration in real-world oceanographic and climate science applications.

Description: This material is based upon work supported by the NSF Graduate Research Fellowship Program and a Microsoft Research PhD Fellowship, as well as the Department of Energy / National Nuclear Security Administration under Award Number DE-NA0003921, the Office of Naval Research under Award Number N00014-17-1-2072, and DARPA under Award Number HR001120C0033.

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Aeronautics and Astronautics at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution February 2023.

Description: Simultaneous localization and mapping (SLAM) is the process by which a robot constructs a global model of an environment from local observations of it; this is a fundamental perceptual capability supporting planning, navigation, and control. We are interested in improving the expressiveness and operational longevity of SLAM systems. In particular, we are interested in leveraging state-of-the-art machine learning methods for object detection to augment the maps robots can build with object-level semantic information. To do so, a robot must combine continuous geometric information about its trajectory and object locations with discrete semantic information about object classes. This problem is complicated by the fact that object detection techniques are often unreliable in novel environments, introducing outliers and making it difficult to determine the correspondence between detected objects and mapped landmarks. For robust long-term navigation, a robot must contend with these discrete sources of ambiguity. Finally, even when measurements are not corrupted by outliers, long-term SLAM remains a challenging computational problem: typical solution methods rely on local optimization techniques that require a good “initial guess,” and whose computational expense grows as measurements accumulate. The first contribution of this thesis addresses the problem of inference for hybrid probabilistic models, i.e., models containing both discrete and continuous states we would like to estimate. These problems frequently arise when modeling e.g., outlier contamination (where binary variables indicate whether a measurement is corrupted), or when performing object-level mapping (where discrete variables may represent measurement-landmark correspondence or object categories). The former application is crucial for designing more robust perception systems. The latter application is especially important for enabling robots to construct semantic maps; that is, maps containing objects whose states are a mixture of continuous (geometric) information and (discrete) categorical information (such as class labels). The second contribution of this thesis is, a novel spectral initialization method which is efficient to compute, easy to implement, and admits the first formal performance guarantees for a SLAM initialization method. The final contribution of this thesis aims to curtail the growing computational expense of long-term SLAM. In particular, we propose an efficient algorithm for graph sparsification capable of reducing the computational burden of SLAM methods without significantly degrading SLAM solution quality. Taken together, these contributions improve the robustness and efficiency of robot perception approaches in the lifelong setting.

Description: This work was generously supported by the NSF Graduate Research Fellowship Program (GRFP), ONR Neuro-Autonomy MURI grant N00014-19-1-2571, ONR grant N00014-18-1-2832, and the MIT-Portugal Program Flagship Project: Knowledge to Data.