Optimized DBSCAN with Improved Static clutter removal for High Resolution Automotive Radars

Density Based Spatial Clustering of Applications with Noise (DBSCAN) is the widely used clustering algorithm for automotive radar applications. Static clutter detections always tend to affect the performance of DBSCAN, especially in high-resolution radar systems. This paper proposes a low complexity static clutter removal method with ego velocity estimation. In addition to the proposed algorithm, this paper discusses a couple of practical problems that are seen with static clutter removed DBSCAN implementations and proposes solutions to them.