Machine Learning and Instrument Autonomy Group


To research, develop, and infuse rigorous solutions built with applied Machine Learning and related technologies to support the exploration of Earth & Space and the advancement of science.

Recent News

✓ The COSMIC team was featured in a Wired magazine article titled "NASA Is Training an AI to Detect Fresh Craters on Mars".
Contributors include (alphabetically) Valentin Bickel (Max Planck Institute for Solar System Research), Ingrid Daubar (JPL/Brown University), Gary Doran (JPL), Annabelle Gao (Brown University), Marko Green (Arizona State University), Lukas Mandrake (JPL), Justin Martia (Arizona State University), Michael Munje (Georgia Institute of Technology), Kiri Wagstaff (JPL), and Daniel Wexler (Brown University).

✓ The COSMIC team recently published a press release on their work finding fresh craters on Mars.
Read the full story here: AI Is Helping Scientists Discover Fresh Craters on Mars

Brian Bue received a 2020 Lew Allen Award for Excellence. This award recognizes early career accomplishments / leadership in scientific research or technological innovation. Dr. Bue was recognized for his contributions to machine learning for remote sensing systems.

Kiri Wagstaff received a 2020 NASA Honor Award Exceptional Technology Achievement Medal for inventing the dynamic landmarking capability (to detect surface features in orbital images) that led to novel onboard mission capabilities and infusion into the Planetary Data System.

Brian Bue is a co-author on a recently published Nature paper titled California’s Methane Super-Emitters.
The full author list is Riley M. Duren, Andrew K. Thorpe, Kelsey T. Foster, Talha Rafiq, Francesca M. Hopkins, Vineet Yadav, Brian D. Bue, David R. Thompson, Stephen Conley, Nadia K. Colombi Christian Frankenberg, Ian B. McCubbin, Michael L. Eastwood, Matthias Falk, Jorn D. Herner, Bart E. Croes, Robert O. Green, and Charles E. Miller. The work is a collaboration between JPL, Caltech, the University of California at Riverside, Harvard University, Scientific Aviation, and the California Air Resources Board. The paper has been reported in several major media outlets, including Discover, Wired, Bloomberg News,, and more.

Infusion Targets

▶ Missions
      ▸ Cubesats: IPEX, NEAScout
      ▸ Rovers: MER, MSL, M2020
      ▸ Deep Space: Europa Clipper, Europa Lander, MRO, Cassini

OCO-2 Spacecraft ▶ Downlinked Datasets
      ▸ Focus of Attention / Content-based Search
      ▸ Validation, Uncertainty / Quality Estimation
      ▸ Science Understanding / Insight Generation
      ▸ Data Exploration & Mining

▶ Real Time / Operational Pipelines
      ▸ Focus of Attention / Rapid Response
      ▸ Triage / Prioritization
      ▸ Content Based Summarization
      ▸ Decision support / Instant Context
      ▸ Quality Estimation

▶ Trusted Autonomy for Science
      ▸ Domain Shift Detection, Characterization, and Correction
      ▸ Interpretable Uncertainty Estimation
      ▸ Explainable Decisions
      ▸ Auto-Calibration
      ▸ Science Utility Estimation
      ▸ Data Summarization to Maximize Science Yield
      ▸ Human-On-The-Loop Systems
      ▸ Reactive Systems

▶ Anomaly Detection / Root Cause Analysis
      ▸ Onboard / Autonomy
      ▸ Triage / Prioritization
      ▸ Knowledge Compression / Summarization
      ▸ Change Detection
Hyperspectral Data Cube

Areas of Expertise / Breadth

▶ Diverse data domains
      ▸ Atmospheric Composition (GCMS)
      ▸ Exoplanet, Atmospheric, Planetary Surface, Near & Deep IR Spectroscopy
      ▸ Orbital (Nadir/Limb) Imagery Earth/Planetary
      ▸ Terrestrial scatterometry / altimetry
      ▸ Orbital/Airborne Synthetic Aperture Radar (SAR)
      ▸ Astrometric, Light Curves
      ▸ Atmospheric/Ionospheric Soundings
      ▸ Hydration via Neutron & Electrochemical Impedance Spectroscopy
      ▸ Magnetometer time series
      ▸ Digital Holographic Microscope movies
      ▸ GPS
      ▸ Langmuir & Plasma Spectrometry
      ▸ Mission telemetry

▶ Data formats
      ▸ Images, Hyperspectral Images, Image Time Series
      ▸ Spectral & Spectrographic
      ▸ High-dimension multivariate time series
      ▸ Geospatial & Spatiotemporal

▶ Technologies & Models
      ▸ Unsupervised (Clustering, Segmentation, Autoencoding, GAN)
      ▸ Supervised (Logistic, Decision Tree, SVM, Random Forest, LSTM, CNN)
      ▸ Classification & Regression
      ▸ Dimensionality Reduction / Visualization (PCA, ICA, MNF, TSNE, SVD)
      ▸ Ensemble (Voting, Boosting)

GEOS-5 Model Assimilating OCO-2 Data

Specialization / Depth

▶ Explainability / Hypothesis Generation
      ▸ Most models built for understanding, not automation
      ▸ Feature selection / alternative feature discovery
      ▸ Provide justification for classification / selection
      ▸ Provide user guidance to help build data intuition
      ▸ Feature engineering to encapsulate learned data structure

▶ Model Uncertainty Quantification
      ▸ Empirical Probability of Correct Classification (not just confidence)
      ▸ Analysis of regions / subsets with high / low performance

▶ Rigorous Verification & Validation
      ▸ Cross-validation of all results, including deep learning
      ▸ Broad sweep hyperparameter optimization
      ▸ Stability analysis

HiRISE images of Seasonal Flows on Warm Martian Slopes ▶ Domain-relevant Impact Metrics
      ▸ Use-case requirement driven
      ▸ Speed of data exploration (hours)
      ▸ Speed of anomaly resolution (hours)
      ▸ New mission capabilities / concepts enabled
      ▸ Reduced data downlink requirements (%)
      ▸ Lowered operational costs ($)
      ▸ Reduced operator skill required

▶ Physical Science Support
      ▸ Publication-supporting product quality
      ▸ Explainable, justifiable model behavior
      ▸ Rigorous performance statistics
      ▸ Intuition and interpretation facilitation
      ▸ Focus-of-attention to permit manual inspection and verification

▶ Model Compression
      ▸ Limited compute environments (onboard autonomy)
      ▸ Low latency / real-time requirements (Ops / autonomy)
      ▸ Multi-model performance comparison / Data insight generation
      ▸ Quick-look capabilities before large analysis computation investment
      ▸ Enhanced explainability / Unnecessary parameter reduction

For more information about working with our group, please contact Lukas Mandrake.

Group Photo