Key Technologies for Intelligent and Safer Cars - from Motion Estimation to Predictive Motion Planning

This paper presents approaches to three of the major challenges for (semi-)autonomous driving in urban environments: self-localization and ego-motion estimation, detection of dynamic objects such as cars and pedestrians, and path planning in dynamic environments. For each of these tasks we present a summary of the techniques we employ and results on real data. All modules have been implemented and tested on our autonomous car platform SmartTer, except from the path planning part which was tested on an indoor robot and is currently being adapted to the SmartTer.