Automated Ground Truth Estimation of Vulnerable Road Users in Automotive Radar Data Using GNSS

Abstract

Annotating automotive radar data is a difficult task. This article presents an automated way of acquiring data labels which uses a highly accurate and portable global navigation satellite system (GNSS). The proposed system is discussed besides a revision of other label acquisitions techniques and a problem description of manual data annotation. The article concludes with a systematic comparison of conventional hand labeling and automatic data acquisition. The results show clear advantages of the proposed method without a relevant loss in labeling accuracy. Minor changes can be observed in the measured radar data, but the so introduced bias of the GNSS reference is clearly outweighed by the indisputable time savings. Beside data annotation, the proposed system can also provide a ground truth for validating object tracking or other automated driving system applications.

Automotive Radar — From First Efforts to Future Systems

“…[137] proposes a semi-automatic labeling based on radar, Lidar, and camera data. In [138], radar measurements are enriched with GNSS ground-truth for ML-based VRU recognition. Furthermore, synthesizing VRU radar responses with radar simulators has been presented with the motion ground-truth obtained from kinematic models [139], animations [140], [141], or Kinect data [142], [143].…”

Automated Ground Truth Estimation for Automotive Radar Tracking Applications With Portable GNSS And IMU Devices

“…According to [3], the proposed reference system consists of the following components: VRUs and vehicle are equipped with a device combining GNSS receiver and an IMU for orientation estimation each. VRUs comprise pedestrians and cyclists for this article.…”

“…Furthermore, both vehicle and VRU can benefit from a position update via IMU if the GNSS signal is erroneous or simply lost for a short period. Experiments in [3] revealed that the standard configuration of the internal position filter, which fuses both signals in the GNSS + IMU unit, is not well equipped for unsteady movements of VRUs, especially not for pedestrians. This quickly led to accumulating positioning errors.…”

“…Label all radar points in ellipses Detectionbased shape refinement φ can be directly adopted from the IMU data. For a stable position filtering, the reference system is also capable to return good estimates for φ and v which makes the more complicated derivation from pure GNSS data in [3] obsolete. The cyclist is labeled inside a rectangle with fixed length of 2.5 m oriented in driving direction φ and width of 1.2 m plus a variable amount based on its yaw rate.…”

“…Let TP (true positives) be the amount of points correctly assigned to a VRU, FP (false positives) the incorrectly assigned points, and FN (false negatives) the amount of points that incorrectly have not been included in a track. Then, the precision of the method can be calcu- precision of 99.48 % and recall of 99.66 % in [3]). Note, that it would certainly be possible to improve these scores for this data set by fine-tuning the parameters of the selection area.…”