Modeling and Validation of Real Time Kinematic Technique in Autonomous Driving Simulator

The ongoing improvement of advanced driving assistance systems (ADAS) is achieved due to the increasing accuracy of sensors of an ego vehicle and due to using vehicle-to-everything (V2X) communication for extra data. In this paper, real time kinematic (RTK) technique that allows to determine high-precision relative position of ego vehicle to other vehicles as well as different infrastructure facilities is presented. The process of developing RTK sensor in autonomous driving simulator CARLA is described. The analysis of the proposed sensor approach, that is based on RTK technique is evaluated in terms of modeling the output data with normal distributed noise. The developed RTK sensor is integrated into an autonomous CARLA simulator. Finally, the simulation results are validated by practical experiment with GNSS equipment and rtklib software.