Welcome to ESSG at MIT!

We develop succinct data-driven models and computational methods for natural systems. We aim to improve the representation, analysis and observation of geophysical and ecological phenomena. Toward this end, we research the classics including stochastic processes, estimation and control, and the modern, including machine learning, and information and pattern theories. You will find that the several projects listed here emphasize the dynamic coupling between phenomenology and physics, simulation and observation, humans and algorithms, and data with models.

Our current contributions include:

  • Algorithms to overcome the curse of nonlinearity, dimensionality and uncertainty in inference problems characteristic of geophysical fluids, particularly by developing a pattern theory of inference for coherent fluids  (stics.mit.edu).
  • Ensemble Learning to mitigate model error effects in nonlinear filtering and smoothing
  • Tractable Information-theoretic learning for autonomous observing systems, integrating data assimilation, uncertainty quantification, adaptive sampling, planning and control for (caos.mit.edu).
  • Developing sUAS to map coherent features (caos.mit.edu). 
  • Incorporating context in Environmental Sensing and Fluid Imaging  (flux.mit.edu).
  • Learning sparse human feedback for high recall in Animal Biometrics (sloop.mit.edu)
  • Physically-based learning  for risk assessment and planning mitigation strategies in a changing climate (crises.mit.edu)

ESSG is strongly interdisciplinary, comfortable developing both methodology and application. You can find us tinkering with "what works in the field," but we are a methodologically inclined group comfortable stretching contours of systems science. At our heart, we are motivated to bring systems engineering to investigations of the earth and environment.

Participants come from many pedagogical areas including EAPS, EECS, Mathematics, and Mechanical and Aerospace Engineering. ESSG is funded in part by the AFOSR Data-Driven Dynamic Application Systems (DDDAS) Program, the National Science Foundation (NSF), Lincoln Laboratory, Naval Undersea Warfare Center Division (NUWC), and MIT's MISTI program.  

Here you will find the people, projects, and publications associated with our group. Enjoy!