AI for Space Application

Increasing the autonomy of space missions.

About the Workshop

Effective use of AI and Machine Learning onboard spacecraft has the possibility of decreasing cost while increasing the value generated from the mission. Towards this goal, AI for Space Application aims to be a cross-disciplinary workshop, bringing together experts in fields such as AI/Machine Learning, Space Plasma Physics, Space Weather, Earth Observation and Spacecraft design and control. The workshop will discuss:

How AI and Machine Learning can be utilized to increase the autonomy of space missions?
What are the challenges?
How can we overcome them?

The workshop is performed as part of the ASAP project, which works to design machine learning algorithms and hardware for use onboard spacecraft. For more information regarding ASAP or the workshop, don't hesitate to get in touch with Jonah Ekelund: https://www.kth.se/profile/jonahek

The Venue

The workshop will take place at Digital Futures Hub, located at the KTH Main Campus in Stockholm, Sweden. For more information on how to get to the venue see: https://www.digitalfutures.kth.se/contact/how-to-get-here/

During the day, fika and lunch will be provided by the Workshop.

Schedule

The workshop will take place on the 28th of January 2025.

09:45–10:00: Welcome
10:00-10:30: Fika
10:30-12:00: Session "Space Weather and Solar Science"
12:00–13:30: Lunch
13:30–15:00: Session "Spacecraft Control and On-board Execution"
15:00–15:30: Fika
15:30-16:00: Discussions
16:00: End

Keynote Speakers

Savvas Raptis

Savvas Raptis is a researcher at the Johns Hopkins University Applied Physics Laboratory. His main expertise lies in multi-point and multi-mission data analysis and in space plasma modeling. His work is focused on using and combining different satellite datasets with computer simulation and machine learning algorithms.

Giacomo Acciarini

Giacomo is completing his PhD at the European Space Agency Advanced Concepts Team, and the University of Surrey. His research focuses on astrodynamics and applied machine learning techniques for space science problems. Since 2020 he has been involved as a faculty in the NASA Frontier Development Lab Program, where he has worked on AI for satellite conjunction avoidance, thermospheric density estimation, onboard atmospheric correction, and more.

Some of his latest publications involve dynamical system theory under uncertainties, spacecraft collision avoidance with probabilistic programming, differentiable programming and orbital propagation, thermospheric density modeling, and space weather forecasting, among other topics.

Sessions

Space Weather and Solar Science

Keynote: Savvas Raptis (Johns Hopkins University Applied Physics Laboratory)

Presentations:

George Miloshevich - "Soft X-ray flux regression via a CNN with weighted loss."
Ekaterina Dineva - "Parametrization of Vector Magnetic Field Images Using Disentangled Representation Learning."
Panagiotis Gonidakis - "Segmentation and Feature Engineering of Solar Coronal Structures Using SDO AIA and HMI Data."

Spacecraft Control and On-board Execution

Keynote: Giacomo Acciarini (European Space Agency - Advanced Concepts Team / University of Surrey)

Presentations:

Jörg Conradt - "Energy-smart Neuromorphic Sensing and Computation for Future Space Applications."
Elias Krantz - "KTH Space Robotics Lab - Simulating Microgravity Robotics for Control and Planning"
Pedro Antunes - "Towards FPGA-Based Neural Network Synthesis for Space Applications."

Automatics in Space Exploration

The principal aim of the ASAP project is to design and develop algorithms for the automation of operations on board space missions based on the use of artificial intelligence techniques to be implemented on the onboard processors. https://asap-space.eu/

Sign-up

Sign-up is open until the 12th 15th of January 2025, confirmation for on-site participation will be sent out before the 15th of January 2025.

Page updated

Google Sites

Report abuse