3-D Object Detection for Multiframe 4-D Automotive Millimeter-Wave Radar Point Cloud

Object detection is a crucial task in autonomous driving. Currently, object-detection methods for autonomous driving systems are primarily based on information from cameras and light detection and ranging (LiDAR), which may experience interference from complex lighting or poor weather. At present, the 4-D (x, y, z, v) millimeter-wave radar can provide a denser point cloud to achieve 3-D object-detection tasks that are difficult to complete with traditional millimeter wave radar. Existing 3-D object point-cloud-detection algorithms are mostly based on 3-D LiDAR; these methods are not necessarily applicable to millimeter-wave radars, which have sparser data and more noise and include velocity information. This study proposes a 3-D object-detection framework based on a multiframe 4-D millimeter-wave radar point cloud. First, the ego vehicle velocity information is estimated by the millimeter-wave radar, and the relative velocity information of the millimeter-wave radar point cloud is compensated for the absolute velocity. Second, by matching between millimeter-wave radar frames, the multiframe millimeter-wave radar point cloud is matched to the last frame. Finally, the object is detected by the proposed multiframe millimeter-wave radar point-cloud-detection network. Experiments are performed using our newly recorded TJ4DRadSet dataset in a complex traffic environment. The results showed that the proposed object-detection framework outperformed the comparison methods based on the 3-D mean average precision. The experimental results and methods can be used as the baseline for other multiframe 4-D millimeter-wave radar-detection algorithms.

Automated summary

This study proposes a 3-D object-detection framework based on a multiframe 4-D millimeter-wave radar point cloud. By matching between millimeter-wave radar frames, the relative velocity information of the point cloud is compensated for the absolute velocity. The proposed framework outperformed the comparison methods based on the 3-D mean average precision.