Blog post – Julia Niebling

 

Julia project imagejulia project image2 greenhouse

Please briefly describe your project exchange participants would be working on and what makes it exciting.

The EDEN ISS project aims at advancing plant cultivation technologies and their application in space, e.g. on-board the International Space Station (ISS) or for future human space exploration missions. Our part in this project is to analyse and use telemetry data collected in the prototype greenhouse located near the Neumayer III Station in Antarctica. As we are at the very beginning of the project, we would like to get a better understanding of that data and gather first insights. After structuring and visualising the data, suitable concrete analysis questions should be formulated and tackled with state-of-the-art methods. Possible applications would be e.g. anomaly detection, prediction or predictive maintenance. The EDEN ISS project does not only play a key role for future long-range space missions but the ability to grow food in extreme environments might also become more relevant on earth.

How did it come to this project? What sparked your interest in it in the first place?

The EDEN ISS project joins forces from different research and industry organisations around the globe to develop bio-regenerative life support systems (BLSS) for incorporation into space stations, transit vehicles and eventually in habitats on planetary outposts. As our working group is concerned with researching machine learning methods and their application within the German Aerospace Center (DLR), we joined the project. The data, collected in the container-sized test-facility greenhouse, offers lots of interesting research questions while we are mainly interested in anomaly detection as a starting point.

Which recent studies from your group were exciting and why?

The project with the closest connection to this one was the recent analysis of railway track irregularities on data collected at the inland harbour of Braunschweig as it also dealt with anomaly detection. Other recent projects are concerned with robust data fusion, the understanding of uncertainties in deep learning and the incorporation of physical knowledge.

What is it like to work in your lab? How many people are there, what is their background?

We have 5-10 people with mainly mathematical or computer science background working in our group, part of whom is located directly at the DLR, some at the “Friedrich-Schiller University” in Jena or the “Technical University of Munich”. Pre Covid-19 most of us have been working in the Institute in Jena with a great view over the nearby “Paradies Park”. At the moment, we all work from home but have weekly meetings to discuss challenges, results and collaboration.

What else would you like to share with data scientists interested in applying?

Whether you already have experience in anomaly detection or not, this project is a great opportunity to apply your knowledge or learn about this interesting research field and explore an early-stage dataset from one of the most unique research projects.