Ego-Motion Estimation and Compensation for Multi-Dimensional Radar Systems on Hovering Platforms

This thesis presents algorithms and concepts for processing signals from multi-dimensional radars on hovering platforms. In particular, the thesis presents methods for locating and monitoring people in search and rescue scenarios using multi-dimensional radars mounted on unmanned aerial vehicles. To enable the flexible adaptation of the presented algorithms, signal definitions for a general multi-dimensional radar are derived.

The main focus of the thesis is the development of algorithms for accurate estimation and compensation of the radar's ego-motion. The radar's ego-motion has to be compensated for, as this is the only way to evaluate the movements of targets, such as people. For this purpose, it is essential to evaluate the phase of the radar signals. In addition, the thesis presents algorithms for the flexible reconstruction of certain geometric areas with a combination of different radar systems as well as methods for extracting and evaluating the chest movement of people. Finally, the necessary extensions of the framework for applications that include scenario changes and translations are proposed.