Exploiting arc splines for digital maps

Abstract

In modern driver assistance systems digital maps have proven usefulness for many applications in providing information on the environment around the host vehicle. For future ADAS, relevant to traffic safety, high precision digital maps including detailed information about the course of individual lanes are needed. We introduce a new method for the generation of highly accurate digital maps including lane specific information based on a wide range of raw data sources. We show that the use of arc splines as a representation model of individual lanes is advantageous for both computational efficiency and accuracy. Therefore, this method is useful for a wide variety of applications using digital maps to enrich driving comfort and traffic safety.