POMDP-based Decision Making for Cognitive Cars using an Adaptive State Space.
Sebastian Klaas · 2011
This thesis analyzes the use of a Partially Observable Markov Decision Process (POMDP) based decision making for cognitive cars. Hence the modeling of vehicles and of the vehicle environment is described (including state-, actionand observation-spaces, transitionand observation-probabilities as well as discounting in infinite-horizon-(PO)MDPs). The focus lies on the combination of continuous and discrete descriptions. The chances and problems of both static and adaptive state spaces (both have been implemented during this work) are being analyzed and discussed. Last but not least the future perspective of this general approach is being evaluated and the work which has to be done until a POMDP-based decision making can be applied to traffic situations in real, complex and highly dynamic environments is outlined.