Sensitivity analysis for model based fusion of camera systems with navigation data

Current advanced driver assistance systems (ADAS) for lateral support are based on optical sensors for environmental detection. Due to different influences e.g. bad sight conditions or missing lane markings, theses systems might not be able determine the necessary information. In order to improve the availability and reliability of camera based ADAS systems in these situations, the potential of a fusion with navigation information is analyzed at the Institut für Kraftfahrzeuge (ika) RWTH Aachen University. Therefore, a fusion model is developed, which continuously fuses data between a camera system with vehicle inertial data and navigation information. In order to gain optimal results out of the fusion model, different sources of potential errors, which have a negative influence on precision, are identified and analyzed within a sensitivity analysis.