In this paper, an explicit decision tree approach for automated driving is proposed. The ego vehicle operates in traffic that is modeled as a discrete-time Markov chain, the state of which is only partially observable. In this setting, the automated driving policy is generated from a decision tree algorithm offline operation to create an explicit implementation for the online use. Simulation results of an implementation of this approach for automated driving on a two-lane highway are reported to illustrate the potential of this approach.

This content is only available via PDF.
You do not currently have access to this content.