Paul Czerwionka

A Three Dimensional Map Format for Autonomous Vehicles

Betreuer: Raúl Rojas , Fritz Ulbrich , Daniel Goehring
Abschluss: Diplom
Abgabedatum: 29.09.2014


Autonomous vehicles of the future need maps with much more information than provided by the existing street maps. State of the art navigation systems already have digital maps, but they lack in detail, accuracy and assume a flat world. Based on the existing two dimensional map format RNDFGraph our autonomous car "MadeInGermany" has been using, this thesis shows the development process and application of a more detailed format covering the third dimension, the Advanced Road Network Definition Graph (ARNDGraph). Over 300 km of Berlin's road network were mapped and autonomously traveled. The experiments include runtime and file size comparisons between the RNDFGraph and the ARNDGraph.