Among the 14 recipients of the MIT School of Science 2016 Infinite Kilometer award was EAPS Principal Research Scientist Sai Ravela.
The MIT School of Science recently announced the 2016 winners of its Infinite Kilometer Award. The Infinite Kilometer Award was established in 2012 to highlight and reward the extraordinary, but often under-recognized work of the school’s research staff and postdocs. Recipients of these awards are exceptional contributors to their research programs. In many cases, they are also deeply committed to their local or global MIT community and are frequently involved in mentoring and advising their junior colleagues, participating in the school’s educational programs, making contributions to the MIT Postdoctoral Association, or contributing to some other facet of the MIT community.
Summarizing Ravela's research interests can be tough. A consummate polymath, he brings his deep knowledge of state estimation to a wide variety of disciplines from electrical engineering to biology.
Perhaps Ravela's most important paper to date, Ravela, Emanual and McLaughlin, 2007, focused on the vexing but enormously important mathematical problem knows as data assimilation, whereby observations of a natural system are combined with prior forecasts of that system using computational models to produce an optimal estimate of the state of the system. This forms the most important and controversial basis for numerical weather prediction, as one example. There are many approaches to the problem, all of which make assumptions about the linearity and Gaussian probability distribution of the errors. This is very problematic in initializing, e.g., numerical weather prediction models because strong, isolated features such as fronts and hurricanes are often mislocated in the prior estimate, resulting in strongly nonlinear and no Gaussian errors. Ravela came up with an ingenious, smooth re-mapping of the prior estimate, now called “field alignment”, which can work in these highly nonlinear environments, and is being adopted by several major forecasting centers. Among the groups interested in applying this new technique is a group at Lincoln Laboratory, with whom Ravela has formed an active collaboration. This group, funded by the FAA, is working on techniques for rapid updates of short-term weather forecasts for the purpose of increasing the efficiency and safety of flight routing.
While continuing to advance the field of state estimation, Ravela has embarked on a number of other ventures. In collaboration with the Emanuel Group, he has worked on the stochastic numerical simulation of tropical cyclone tracks, and developed numerically advanced hydrodynamic surge models. He has also worked on advancing the field of signal detection. On the side, he designed and built robotic aircraft (UAVs) which he successfully flew into the volcanic plume of the Mexican volcano Popocatepetl, measuring the chemical composition of the plume.
Ravela is also an effective educator. He has co-supervised several graduate students (his position did not allow him to be a primary supervisor) as well as several interns, all of whom have profited greatly from his advice and mentorship. He has also co-taught several advanced courses and led multiple IAP activities. On top of all of this, he is involved in industry, acting as a consultant and also as a partner in a new risk assessment firm. Not surprisingly, in the process, he has made strong connections across EAPS, in both the Schools of Science and Engineering, and indeed MIT as a whole, forming collaborations with groups at Lincoln Labs and even in the Sloan School.
Beyond his research and teaching, this award also recognizes Ravela as an excellent “MIT citizen”, who has helped teach advanced courses and lead IAP activities. A committed long time environmentalist, he currently serves on MIT’s Campus Resilience Committee where he is actively involved in using advanced models to estimate MIT’s own risks stemming from sea level rise and climate change. He is also a good regional citizen, serving on Boston’s Green Ribbon Committee to help assess flood risk to Back Bay Boston.
In addition to an award of $1,000, honorees and their colleagues, friends, and family will be invited to an award lunch this spring.
Story picture: Field tests to fly an AUV into the plume from Popocatépetl (right) - Image courtesy: caos.mit.edu
Principal Research Scientist Sai Ravela studies estimation, control and information theories, and topics in statistical pattern recognition and statistical inference and learning with the goal of designing methods that succinctly represent stochastic signals and systems.