Eric Robinson singled out for outstanding performance in the field of science and engineering
CAMBRIDGE, MA—The burgeoning commercial space industry has attracted billions of dollars in investment. Upwards of 1,000 satellites circle the Earth, beaming down data and images every minute. All of this technology promises to let astronauts, businesses and space tourists alike gather more information from and about space than ever before.
But according to Eric Robinson, a Mobility Pilot in the U.S. Air Force, there may be an unintended consequence to all of this discovery. “There’s going to be a huge and growing demand for satellite data,” he said, “but mechanisms for coordinating data gathering among satellites, which would vastly increase the utility that can be gained from these assets, are nonexistent.” To put it in industry parlance: Houston, we’ve got a data problem.
But for Robinson, the promise of richer data presents an opportunity. He found a way to optimize the tasking of data collection across separately managed satellites. His method overcomes the current practice of keeping information “stovepiped” and out of the hands of others. “Even within organizations, including branches of the Department of Defense and NASA, there may be separate, generally unshared, collection assets. That creates inefficiencies that could be avoided through coordinated planning,” Robinson said.
To improve data gathering, Robinson developed a set of algorithms that taps into the satellite schedule planners and builds a way to ensure the “asks” for data collection and allocation are efficient and optimized, unlike the manual intervention that would be required to coordinate that today. He used machine learning to automatically find patterns and structures in the satellite data in order to identify hidden uncertainties and the information needed to complete an imagery collect by a given satellite.
In recognition of his achievement, the Air Force Association honored Robinson with its Theodore Von Karman Award during a recent ceremony at the Air Force Association’s Air, Space & Cyber Conference.
One of the potential payoffs for all of this data sharing could be better weather forecasting, according to Draper engineer Mark Abramson. “Predicting and managing the response to severe weather events, like hurricanes, gets easier when you have a system that gives you simultaneous measurements, persistent surveillance, minimal information gaps and soonest-possible observations. By linking satellites in a web of information, we can help improve the accuracy of, for example, how scientists monitor the ozone levels near the poles, measure atmospheric and oceanic properties and examine both emitted and reflected radiation from Earth’s surface.”
Robinson’s research was conducted while he was a member of the Draper Fellow Program under the guidance of Abramson and Stephen Kolitz, also of Draper, and as a graduate student in the MIT Aeronautical and Astrophysical department under Associate Professor Hamsa Balakrishnan. Since the 1970s, Draper has guided and supported more than 1,200 Draper Fellows, who have come from both civilian and military backgrounds and contribute all over the world in technical, corporate, government, academia and entrepreneurship sectors. Draper Fellow alumni have gone on to serve as high-ranking members in the DoD, astronauts for NASA, leaders in industry and faculty members in engineering and the sciences.
Robinson’s work continues Draper’s decades-long support of NASA, including support for NASA’s Hurricane and Severe Storm Sentinel (HS3) mission and NASA’s Earth Science Technology Office’s Advanced Information System Technology program. Robinson’s thesis work was completed under Draper’s fourth NASA-funded AIST project.
Draper develops novel PN&T solutions by combining precision instrumentation, advanced hardware technology, comprehensive algorithm and software development skills, and unique infrastructure and test resources to deploy system solutions. The scope of these efforts generally focuses on guidance, navigation, and control GN&C-related needs, ranging from highly accurate, inertial solutions for (ICBMs) and inertial/stellar solutions for SLBMs, to integrated Inertial Navigation System(INS)/GPS solutions for gun-fired munitions, to multisensor configurations for soldier navigation in GPS-challenged environments. Emerging technologies under development that leverage and advance commercial technology offerings include celestial navigation (compact star cameras), inertial navigation (MEMS, cold atom sensors), precision time transfer (precision optics, chip-scale atomic clocks) and vision-based navigation (cell phone cameras, combinatorial signal processing algorithms).
Draper combines specific domain expertise and knowledge of how to apply the latest analytics techniques to extract meaningful information from raw data to better understand complex, dynamic processes. Our system design approach encompasses effective organization and processing of large data sets, automated analysis using algorithms and exploitation of results. To facilitate user interaction with these processed data sets, Draper applies advanced techniques to automate understanding and correlation of patterns in the data. Draper’s expertise encompasses machine learning (including deep learning), information fusion from diverse and heterogeneous data sources, optimized coupling of data acquisition and analysis and novel methods for analysis of imagery and video data.