Machine Learning and Instrument Autonomy Group

We aim to measurably improve JPL’s scientific understanding and productivity within Astrophysics, Earth and Planetary sciences by infusing data science techniques, such as uncertainty quantification, model-based inference, machine learning, and artificial intelligence, that can directly support science and its applications. The council shall develop a strategy that advances JPL leadership, in collaboration with Campus, in scientific understanding based on state-of-the-art data science methodologies. This council will first evaluate the current state of practice at JPL for scientific understanding from data science methods, characterize past successful science/data science collaborations, identify several high priority outstanding problems with the highest potential benefit from state-of-the-art data science methods, and consider needed organizational or culture changes, specific research directions, staffing issues, and beneficial partnerships with academia, other Federally Funded Research and Development Centers, and the commercial sector. The recommendations should explicitly tie solutions and changes to measurable benefits to JPL science and contain timelines, success criteria, and required resources. Outcomes will include a roadmap for future activities together with approaches to formulate science questions for use with traditional and new data science methods.

In order to preserve focus on direct JPL science impact, the council will not consider the following data science application areas: engineering telemetry and modeling, onboard autonomy, intelligent instruments, mission data systems, planning/scheduling systems, or business & HR support.

Machine learning and signal processing methods offer significant benefits to the geosciences, but realizing this potential will require closer engagement among different research communities (article link).

Machine learning is gaining popularity across scientific and technical fields, but it’s often not clear to researchers, especially young scientists, how they can apply these methods in their work (article link).

Overview of data science for physical scientists by Dr. Lukas Mandrake at the JPL SUDS Seminar Series