RCS Estimation and Synthesis of Typical Traffic Participants

Abstract

For a road scenario simulation of automotive millimeter-wave radar, the properties of the radar cross section (RCS) of typical traffic participants will greatly determine the quality of reproduction of the real environment. In the field of road scenario simulation test of advanced driving assistance system (ADAS), it is common to measure the RCS in an anechoic chamber by using vector network analyzer (VNA) to generate RCS models of typical traffic participants. However, due to the strong fluctuation of the targets, the accuracy is low when the RCS models are directly used in the simulation test. On the other hand, the existence of multi-scattering centers of complex targets makes it difficult to use a single point to represent the RCS properties of typical traffic participants which causes errors compared to the real targets. In this paper a RCS estimation and synthesis method of typical traffic participants is proposed to improve the accuracy of target simulation. This method takes global vehicle target (GVT), pedestrian target and cyclist target as examples, which are used commonly in road test of vehicle ADAS. First, the RCS of the targets mentioned above are measures in an anechoic chamber as the targets are rotated at various angles with a VNA and then the RCS characteristics of targets versus the rotating angle can be plotted and analyzed. Then, according to the characteristics of multi-scattering centers of complex target, the distribution and intensity of the scattering centers of various typical traffic participants in different positions and attitudes are estimated. The strong scattering centers of the target are extracted and later synthesized to realize the simulation of traffic participants and improve the simulation accuracy.