JOB DESCRIPTION:

In order to support the advancement in the field of conjunction analysis and collision avoidance of space assets, Airbus Defence and Space is looking for a

PHD STUDENT IN THE FIELD OF ON-BOARD COLLISION AVOIDANCE (D/F/M)

You are looking for a PhD and want to get to know the work of a job title? Then apply now! We look forward to you supporting us in the TSEAF1 department as a Doctorand (d/f/m)!

• Location: Friedrichshafen
• Start: 01.10.2025
• Duration: 3 years

DESIRED SKILLS AND QUALIFICATIONS

• Masters’ Degree in the area of Space Engineering, Electrical Engineering, Computer Science, Mathematics or an equivalent field of study.
• Experience with embedded programming, ideally VHDL know-how.
• Experience with AI algorithms (ideally machine learning and/or deep learning) and their development/training.
• Experience with programming languages like Python, C++ or C
• Advanced understanding of orbital guidance, navigation, control and related processes (orbit propagation, orbit determination, etc.) would be a key asset.
• Experience with software Tools like Matlab would be a plus.
• Experience with data exchange protocols (e.g. TCP/IP or similar) would be a plus.
• Experience with Real-Time Operating Systems (e.g. RTEMs or similar) would be a plus.
• Fluent in English, fluency in German would be a plus.
• Ability to work independently and to solve complex problems.
• Please upload the following documents: cover letter, CV, relevant transcripts, masters’ degree certificate.
  Not a 100% match? No worries! Airbus supports your personal growth.
  Take your career to a new level and apply online now!
  This job requires an awareness of any potential compliance risks and a commitment to act with integrity, as the foundation for the Company’s success, reputation and sustainable growth.

EXPERIENCE LEVEL:

Student

Collision Avoidance is a crucial element of satellite operations. The collision avoidance mission segment (infrastructure and tools) is considered mission critical, as close encounters with either space debris or actively controlled satellites are common for every mission.
The research objective of the proposed PhD is to research on the existing challenges and to propose and deploy an embedded on-board end to end application for autonomous collision avoidance on a flight representative processor in the loop (PiL) bench. The PiL shall be integrated within a higher-level system simulator in order to demonstrate the system capabilities in a simulated scenario. applications.

Ability to identify close approaches

• Identification and Implementation of conjunction screening algorithms compatible with on-board hardware
• Data input definition (e.g. orbital catalogs) such that the amount of data to be uploaded is compatible with traditional S-Band uplink rates and volumes without compromising safety
• Specification of TM/TC interface (requests and reports for Collision Avoidance-related tasks (e.g. “start screening”, “stop screening”, “report results”, etc.)

• Validation on hardware, representative of the performance of flight-proven on-board computers

  Maneuverability detection using autonomous adavanced algorithms (for example, machine learning, deep learning etc..)

• Apart from few CubeSats, most objects orbiting Earth which are actively controlled by ground are also capable maneuvering (chemical, electrical or differential drag maneuvers). It is important to identify such objects such that the owners/operators can coordinate encounters of actively controlled objects. Should the secondary object not be able to maneuver, an autonomous derisking sequence can be safely initiated. The autonomous identification of maneuverability is challenging. Traditional, i.e. non-AI methods have often failed in the past, as the separation of natural orbital disturbances from actual maneuvers is non-trivial, especially if the maneuvers are small in thrust or only rarely executed. As part of the proposed PhD studies, Artificial Intelligence shall be investigated as technology to achieving robust maneuverability estimations. The developed solution shall then be deployed on ”flight representative” hardware bench for validation of potential future flight eligibility.

  Assisted Collision Risk Assessment

• Should the secondary object not be capable of maneuvering, an autonomous derisking sequence can be initiated. This however requires the ability of collision risk assessment. Traditionally, collision risk assessment is based on the computation of various kinds of collision probabilities (“Pc”, “PcMax”, etc.). Oftentimes, the data available to specialized ground personnel is not conclusive or of bad quality, which impacts the decision-making process. As part of the proposed PhD studies, it shall be investigated, to which extent novel solutions (for example AI) can be used to improve this process. Multiple paths need to be investigated, for example, but not limited to:

• Not following the traditional path – e.g. instead of computing collision probabilities, etc. past encounters and their maneuvering decisions could be used for training of a decision-making AI

• Following the traditional path, but integrating AI into the decision-making process, e.g. employing AI to detect so-called “Dilution region”-encounters, detect low relative velocity encounters/”Long-Term Encounter”, correct orbital uncertainty information of bad quality

  Autonomous derisking

• Based on simple thresholding and standardization of maneuver strategies, a collision avoidance maneuver can be computed. This maneuver then needs to be cross-checked against the most recent knowledge of the orbital neighborhood in order to ensure that no secondary conjunctions are created through the collision avoidance maneuver. If potential secondary conjunctions are identified and deemed of high risk (exercising again steps 2 and 3 of the pipeline), the maneuver needs to be adapted until a safe strategy is found. After derisking, a maneuver sequence shall be computed to safely and swiftly return to the nominal orbit.