Parallelised Least Squares Adjustment (Internship or Thesis)


The Delft Method for adjustment and testing is a rigorous framework to achieve the best unbiased estimation of coordinates from observations. The pinnacle of this method is least squares adjustment. In recent years, the number of available observations has grown exponentially, making least squares adjustment the bottleneck in data processing. In your project you will be doing fundamental research on parallelising least squares adjustment.

What you will do

In your project you will investigate the current least squares adjustment applications in use at Geodelta. You will also do literature research on some existing parallelisation methods for least squares adjustment. Based on your research you will investigate the suitability of these existing methods for application at Geodelta’s computations. Preferably you will also work on an implementation of this method as a proof-of-concept.

Your background

For this position we are looking for an intern or thesis student. You are currently studying Mathematics, Geodesy or Geoscience & Remote Sensing. You have a good understanding of least squares adjustment and linear algebra.

What we offer

A challenging and innovative project on the edge of technology with immediate impact in real-life applications. You will work from the office of Geodelta, which is based in the city center of Delft. Geodelta offers a monthly internship compensation for the duration of your project.

More information

To apply for this position, or for more information, contact Martin Kodde ( You can also review our other vacancies.