Generation of 3D Grid-Maps Using an Incoherent Network of Radar Sensors with 1D Angle Estimation

Radar sensors are nowadays often used in modern driver assistance systems to generate robust and precise representations of the environment for example with the help of grid maps. 3D environmental maps offer the advantage of not only identifying occupied and unoccupied areas, but also providing height information directly. However, the generation of such maps require a radar sensor that can determine the position of objects in the 3D space. In this work, two algorithms are presented that creates a 3D representation of the environment based solely on incoherently networked multiple input multiple output (MIMO) radar sensors with 1D angle estimation. Measurements show that the presented algorithm provides precise 3D target localization with a maximum standard deviation of 6 cm for occupied cells. This is shown for both point targets and extended targets, such as bridges which are of most interest in automotive scenarios.