[Robotics] Master’s thesis: Robot localization with radar: Measurement model based on full radar images

15.01.2025, Abschlussarbeiten, Bachelor- und Masterarbeiten

Spinning radar sensors are emerging as an interesting sensing modality for localization and mapping in robotics. They offer full spectrum measurements of the environment but are subject to environment-specific multipath errors. Most radar localization frameworks today aim to suppress the noise by converting the dense radar spectrum measurements to a sparse lidar-like point cloud. Recent advances [1] enable simulation of the full radar spectrum for multiple possible robot locations, which paves the way towards Monte-Carlo Localization [2] of robots using radar.
The goal of this project is to develop a measurement model for radar utilizing the full returned radar spectrum. This model will be compared to a baseline implementation based on existing lidar measurement models. In the end, it should be investigated whether the use of multipath information can improve the measurement model.

Possible work packages are:
- Literature review of measurement models for radar sensors
- Implementation of different measurement models in a simulated environment
- Comparison of the measurement models in simulation
- Validation using a real radar sensor

Prerequisites:
- Programming experience in Python and C++
- Student at CIT
- Experience with ROS, Gazebo, and Docker preferable
- Knowledge of robotics
- Independent and self-sufficient way of working

The thesis is supervised by the chair “Perception for Intelligent Systems” (Prof. Lilienthal). We are a new, growing chair with a flat hierarchy and many national and international partners. You can find more information about our research at https://www.ce.cit.tum.de/pins/startseite/

References:
[1]: Mock, Alexander, Martin Magnusson, and Joachim Hertzberg. "RadaRays: Real-time Simulation of Rotating FMCW Radar for Mobile Robotics via Hardware-accelerated Ray Tracing." arXiv preprint arXiv:2310.03505 (2023).
[2]: Dellaert, Frank, et al. "Monte carlo localization for mobile robots." Proceedings 1999 IEEE international conference on robotics and automation (Cat. No. 99CH36288C). Vol. 2. IEEE, 1999.

Kontakt: maximilian.hilger@tum.de