AI for space object orbit predictions

Functie

Background
With the decrease in satellite sizes and increase in number of launches and space debris around Earth, it is becoming more and more important to avoid collisions. A frequently used model used for the propagation of space object orbits is SGP4, which takes into account various influences but is far from complete. Currently, because of the resulting prediction uncertainties, a warning is given a few days in advance of a possible collision with a threshold on chance of occurence. The majority of evasive maneuvers however is proven unnecessary after the conjunction event which limits the mission time of the evading satellite. An improvement of the orbit prediction method is being researched by many companies and space organizations, and NLR is researching the application of AI as an approach.

Assignment
The objective of the assignment is to develop a AI algorithm which has a better prediction for space object orbits as compared to the current SGP4 model. This is based on preliminary research which determined that using AI for this is actually possible and presented some first recommendations and code examples. The topic is further being researched at NLR, and the student will be joining the bigger Space Situational Awareness team and collaborate directly with NLR employees working on the topic. Assistance by the AI experts at NLR is available and foreseen as well. A space object database, including all Low Earth Orbit observed objects, is being obtained daily while single object historical data can be requested.

Tasks to be foreseen in the project:

• Further investigation of available and usuable AI models
• Applying smart filtering trade-offs to determine dataset limitations necessary
• (Assist in) Training of the AI algorithms
• Testing the AI algorithms with the data available to do a performance evaluation
• Documenting the research in an NLR report

Result
The following outcome of the internship is expected:

• One or more AI algorithms for improved propagation of space objects
• Recommended developments to further improve accuracy
• An internship technical report

Duration
3-6 months (internship)

Profiel

Requested

• 4th year master student in the following fields:
  • Information technology
  • Geo-information sciences & Earth observation
  • Engineering & technology
  • Data sciences
  • Aerospace engineering
• Interest in software development and AI, understanding of datasets
• Coding experience (preferably Python/Matlab/C++)
• Experience in astrodynamics and differential calculus is a plus
• Assertive and self-motivated, able to be part of the project team and also proceed individually

Arbeidsvoorwaarden

What we offer

• A challenging graduation project/ internship in a high-tech result orientated work environment
• Weekly supervision and availability of the technical staff for support
• An internship allowance
• Working in an actual R&D project as part of the team
• Internship results to be used in the current and future projects

Informatie

About NLR
The Royal Netherlands Aerospace Centre NLR is the research organization in the Netherlands in the field of aerospace engineering. Around 650 employees highly educated from aircraft engineers to psychologists and from mathematicians to application experts. Visit our NLR media channel on YouTube for a good impression of the organization.

This assignment will be mainly managed by the AI group within the Aerospace Systems and Information Supremacy (ASIS) department.

Solliciteren

Send your application, together with your motivation letter and CV to Alexander.Haagsma@nlr.nl. Phone number: +31(0)88 511 33 94. A first selection of candidates will be made ASAP. However more internships related to, and in continuation of, the current one are foreseen. Therefore late responses are also appreciated.