Instantaneous Ego-Motion Estimation using Doppler Radar

摘要

The growing use of Doppler radars in the automotive field and the constantly increasing measurement accuracy open new possibilities for estimating the motion of the ego-vehicle. The following paper presents a robust and self-contained algorithm to instantly determine the velocity and yaw rate of the ego-vehicle. The algorithm is based on the received reflections (targets) of a single measurement cycle. It analyzes the distribution of their radial velocities over the azimuth angle. The algorithm does not require any preprocessing steps such as clustering or clutter suppression. Storage of history and data association is avoided. As an additional benefit, all targets are instantly labeled as stationary or non-stationary.