Learning to ground fact symbols in behavior-based robots

A robot running a hybrid control system (its architecture comprising a deliberative and a reactive part) must permanently update its symbolic situation model to allow its ongoing deliberation to operate. Previous work has shown that this update can be improved by using, possibly among other sources, the robot's sensor information as filtered through recent activation value histories of robot behaviors. In that work, characteristic patterns in groups of behavior activation values are used to define chronicles, which allow true facts about the current situation to be hypothesized. Chronicle definitions are hand-crafted as part of the domain modeling. In this paper, we demonstrate that analogs of chronicle definitions can be learned. We use an echo state network, which is a particular type of recurrent neural network. To train it, the same activation value data are used as before in chronicle definitions. The training process is fast. The detection process is cheap to run on-line on board the robot. The new method is demonstrated on data from a robot simulator. It provides the robot programmer an alternative tool for getting recent symbolic situation fact hypotheses.