Improving Height Estimation for Stationary Targets with 3D Automotive Radar: From Uncertainty Analysis to Temporal Filtering
Chun-Yu Hou, Chieh‐Chih Wang, Wen‐Chieh Lin · 2025
This paper addresses the challenge of elevation angle estimation in 3D automotive radar, a critical limitation for achieving accurate and reliable 3D scene understanding in autonomous driving. While vertical Doppler Beam Sharpening (DBS) provides a foundation for height estimation, existing implementations often suffer from limitations due to measurement noise. We enhance DBS using a rigorous uncertainty analysis and a robust, temporal filtering approach. Our analysis reveals the significant impact of target-sensor geometry, particularly small elevation angles, on estimation errors. To mitigate these uncertainties, we develop a simple yet effective method combining an Extended Kalman Filter (EKF) for temporal filtering with robust data association to reject spurious detections. Real-world experiments on highway and ITRI campus datasets, spanning 34 km and 1.9 km respectively, using a standard 3D radar and a prebuilt LiDAR map for ground truth, demonstrate a substantial improvement in height accuracy. Compared to unfiltered DBS, our method increases height accuracy within 1 meter from 53.41% to 62.32% on the highway and from 47.74% to 57.56% on ITRI campus.