Engineer (f/m/d) - Error modelling and spatial AI for safe multisensor navigation
Vacancy-ID: 5546 Place of work: Oberpfaffenhofen Starting date: as soon as possible Career level: Experienced professionals; Graduates Type of employment: Full-time; Part time Duration of contract: 2 years Remuneration: Remuneration is in accordance with the Collective Agreement for the Public Sector - Federal Government (TVöD-Bund) The Institute of Communications and Navigation, with sites in Oberpfaffenhofen and Neustrelitz, is dedicated to mission-oriented research in selcted areas of communications and navigation.
These are embedded in the DLR programmes on space, aviation, transport and security. What to expect In contrast to the techniques currently in commercial use, the Institute’s research focuses on ensuring the reliability and integrity of the position information determined.The underlying idea is that fully autonomous driving or flying is only possible if it can be demonstrated that multisensor navigation is safe within the specified parameters. To this end, it is necessary to analyse how different types of measurement errors affect the system and how these propagate statistically through the various processing stages.
This is very challenging for the neural networks typically used today, so new approaches are to be developed, analysed and evaluated. You will be expected to contribute to the development and analysis of reliable methods for visual navigation and sensor fusion within the aforementioned overarching research area.
Your tasks Design and development of innovative algorithms to improve multisensor navigation, meeting stringent safety requirements for road, rail and air transport Development of error models to quantify the integrity of multisensor navigation methods, e.g. calibration error and detection error Creation of realistic simulation scenarios using 3D modelling tools, e.g. game engines, to generate training data and test the developed methods Development and implementation of a suitable simulation and evaluation environment using programming languages such as C/C++ and Python Analysis of the simulation results obtained to assess error propagation and error probabilities Validation of the developed methods using hardware data collected during measurement campaigns Your profile A completed university degree (Master’s / University Diploma) in engineering (e.g. Computer Science, Aerospace Engineering, Electrical Engineering and Information Technology) or other degrees relevant to the role Solid foundations in mathematics and statistics Knowledge of multisensor navigation, visual navigation and machine learning Knowledge of the application of Vision Transformers for image perception Good knowledge of the programming languages Python and C/C++ Very good English skill Remuneration depends on qualifications and job assignment up to pay grad 13 TVöD.
We look forward to getting to know you! If you have any questions about this position (Vacancy-ID 5546 ) please contact: Prof. Dr. Michael Meurer Tel.: +49 8153 28 3065