The Machine Learning and Instrument Autonomy strives to research, develop, and infuse data driven solutions built with applied Machine Learning and related technologies to support the exploration of Earth & Space and the advancement of science. MLIA is enabling data science approach for JPL’s past and future missions, such as OCO-2/3, SBG, Europa Clipper, NISR, and M2020.
Focus Areas
| Instrument Autonomy | Science Serving Data Science |
|---|---|
| Content-based Search | Science Understanding / Insight Generation |
| Data Triage and Down-link Prioritization | Interpretable Models |
| Knowledge Compression and Summarization | ML Uncertainty Awareness |
| Change and Novelty Detection | Data Exploration & Discovery |
How We Work
MLIA works in tandem with instrument and mission PI’s and scientists across JPL. Typically, collaborations start with short, non-committal brain storming sessions to explore new challenges, determine whether data-driven solutions would be appropriate, and recommend approaches with the most promise . Next, we partner to draft a proposal targeting either internal (SRTD, TRTD, JNEXT) or external (ROSES, AIST, ACCESS) funding opportunities. Once funding is secured, collaborative research can begin. Together with our partners, we use data science approaches to explore and understand the data, develop simple interpretable models, and generate insight. As needed and only as much as necessary, we incorporate increasingly sophisticated technology to achieve the project goals. However, even when using cutting-edge approaches like deep learning, we strive to extract as much insight as feasible into the original data, the behavior and capability of the model, and the key drivers of uncertainty and bias.
Contact
Want to collaborate or do some quick brainstorming together?
Want to join our team? We are looking for new colleagues that know how to generate insight from complex data, have experience with UQ, or demonstrated how to merge physical and statistical models.
Please reach out to Dr. Umaa Rebbapragada.
Recent News
-
▸ April 11th, 2024 by Mario Damiano
Dr. Mario Damiano was invited to be members of the Steering Committee (SC) of the Habitable Worlds Observatory (HWO) Characterizing Exoplanet Working Group (WG). In addition, Mario Damiano was invited to co-lead the development of science case(s) around the detection and characterization of water worlds. The Characterizing Exoplanet WG charter is to “Investigate HWO science cases related to remote sensing observations of exoplanets using high-contrast imaging and UV/optical/near-infrared spectroscopy, as well as transit spectroscopy.”
-
▸ April 9th, 2024 by Ryan McGranaghan
Dr. Ryan McGranaghan gave the Early Career Keynote Presentation at the American Geophysical Union (AGU) Triennial Earth-Sun Summit entitled, “Complexity Heliophysics: A lived and living history and its grand opportunity.” The talk presented the lived and living history of complexity science as a paradigm of scientific discovery in Earth and space science, arriving at three pathways that this paradigm makes clear are a future for these fields: 1) Complexity – AI: understanding the connections between complexity science and machine learning/artificial intelligence; 2) Risk science and resilience: A ‘risk science’ framework can be a layer where fundamental understanding and predictive capability converge; and 3) Cultural challenges: interdisciplinary work requires new capacities for facilitation and team composition as well as new capabilities for integrating information, which together require new ways of ‘looking’ at a community (e.g., indicators and metrics).
-
▸ March 29th, 2024 by Philip Brodrick
The ISOFIT team released version 3.0, a major update that facilitates new advances for the next generation of imaging spectroscopy retrievals. ISOFIT is the surface and atmospheric modeling optimal estimation codebase utilized for the AVIRIS programs, EMIT, Carbon Mapper, SBG, and more. Led by JPL, the open-source effort has use and engagement by industry, international partners, and other NASA centers. James Montgomery (398J) contributed large amounts of the development, and the project is led by Phil Brodrick (398J) and David Thompson (382B) - however many members of the JPL community actively contribute.
-
▸ January 30th, 2024 by Mark Wronkiewicz
The OWLS-Autonomy team published a paper on their science autonomy work completed under JNEXT’s Ocean Worlds Life Surveyor (OWLS) project. Group members included Mark Wronkiewicz, Jake Lee, Jack Lightholder, and Steffen Mauceri. The OWLS-Autonomy publication marks a significant advancement in science autonomy, showcasing how onboard algorithms can improve science outcomes on future life detection missions to the outer solar system. It details the development of two Onboard Science Instrument Autonomy (OSIA) algorithms to summarize and prioritize instrument data with Snapdragon-level compute. The paper also highlights key development principles and lessons learned, including notional flight-like requirements and the value of integrating early with science and hardware teams to build trust. Finally, it includes results from a week-long field test of the integrated OWLS platform at Mono Lake, CA, resulting in a TRL 5 classification of the OSIA software.