Physicist, data or computer scientist

The Institute for Optical Sensor Systems (OS) develops, builds, verifies, and operates optical sensors for various tasks. In the department "In-Situ Sensing" (ISS), optical systems are developed and methods are investigated which are suitable for the in-situ exploration of planetary surfaces. Spectroscopy methods such as laser-induced plasma spectroscopy (LIBS) or Raman spectroscopy are already in operation on Mars, for example, by NASAs Curiosity and Perseverance rovers. In the DLR junior research group "Machine Learning for Planetary in-situ Spectroscopic Data", the name is the program, we explore machine learning (ML) methods for enhancing the efficiency of such spectroscopic data analysis. We are a young team involved in NASAs missions working on data taken in-situ on Mars. Furthermore, we measure data in the laboratory and during field campaigns focusing on conditions not only relevant for Mars but also for other Solar System bodies such as the Moon and asteroids.

Ihre Aufgaben

Type of employment: Part-time You will work 19.25 hours per week.

PhD position

Your activity:

Your tasks will include:

• conducting measurements in the laboratory with highly performant but also compact instrumentation to generate training data but also to compare real spectra with predicted spectra

• comparing the performance of different types of machine learning approaches, e.g. neural network architectures

• acquire in-depth knowledge of the physics of the spectroscopy methods and use this for the implementation of SciML

• publishing scientific papers and giving presentations at national and international conferences

• contribution to national and international collaborations

Ihr Profil

The physical processes behind spectroscopy are often complex especially with regard to geological samples and changing experimental conditions which complicates quantitative predictions and unambiguous classification. Therefore, more and more machine learning algorithms are used to potentially reduce uncertainties. Following recent trends of scientific machine learning (SciML), physical knowledge of the measurement method or of experimental factors should be incorporated into the training. For the junior research group, we are looking for a highly motivated and enthusiastic PhD student with a strong interest in combining two exciting fields of research: Solar System exploration and data science. Your mission will be to address the complexity of spectroscopic data with methods from the field of machine learning.,

• Master in physics, computational or data sciences, geoinformatics or a similar field

• very good expertise in programming (e.g. Python) and in the application of methods from data science

• the Solar System fascinates you and you want to contribute to its exploration.

• good self-organization skills

• high degree of initiative in tackling complex problems

• You collaborate well in a team environment.

• very good communication skills in English