A Moving Obstacles Detection Method Based on Millimeter-Wave Radar

An increasing amount of research is dedicated to ensuring train safety through obstacle detection. This study focuses on addressing dynamic targets in front of trains, as they pose higher uncertainty and potential risks compared to static targets. The proposed approach leverages the integration of millimeter wave radar with the DBSCAN clustering algorithm and velocity filtering algorithm to achieve accurate and reliable detection of dynamic targets along the train's path. Practical experiments were conducted to evaluate the effectiveness of the proposed approach. The results demonstrated an impressive 90% accuracy rate when the millimeter wave radar was in a state of constant motion, showcasing the proficiency of the DBSCAN and velocity filtering algorithm in identifying genuine dynamic targets while effectively filtering out false positives. Moreover, compared to the Kalman filtering algorithm, the processing speed was significantly faster, taking only a few milliseconds. These research findings contribute to enhancing train operation safety through improved dynamic target detection.