While more and more advanced driver assistance systems (ADAS) are getting on the market and people already start dreaming of fully automated (autonomous) driving (AD), we have to ask the important question how those automated driving functions will get tested and validated in order to proof it safe.
Therefore, JOANNEUM RESEARCH is producing Ultra High Definition Maps (UHDmaps) to pave the ground for test and validation of AD functions by simulation, real-test and test-beds. UHDmaps contain a digital copy of reality, which:
- enables realistic simulation;
- sets the benchmark for the assessment of automated driving functions in real-tests;
- enables sensor stimulation in test-bed environments by providing real Radar and Lidar reflections mapped to the surfaces of the textured 3D model.
The automatic creation of UHDmaps is one of the challenges JOANNEUM RESEARCH is facing. Our scalable and cost efficient approach is based on mobile mapping data captured with a “Leica Pegasus 2 Ultimate Dual Head” (funded by “Zukunftsfond Steiermark”) which is fed into an AI driven workflow to deduce map features automatically. Thus, manual work is limited to quality control only.
In order to develop and evaluate future proof technologies and solution for the creation of UHDmaps JOANNEUM RESEARCH got strongly involved in the Austrian Light Vehicle Proving Region for Automated Driving (ALP.Lab GmbH.) as a shareholder and a scientific partner. Test and validation of automated driving functions are carried out by ALP.Lab partners in simulation, test-bed and real-test based on UHDmaps. This generates valuable feedback regarding future scientific requirements of the automotive industry in the challenging field of test and validation of automated driving functions.