Logical Verification of AND/OR-net Structures for Task-Knowledge Representation in Service Robotic Scenarios

Logical Verification of AND/OR-net Structures for Task-Knowledge Representation in Service Robotic Scenarios

  Service robots and especially rehabilitation robots shall support disabled persons in daily life situations as well as in the working environment. Currently available systems offer support on a relatively low task level. This puts a high cognitive load on its users, so that controlling these systems turns out to be very tiresome. In order to eliminate this disadvantage system control on a higher task level becomes a necessary prerequisite. For this purpose, detailed information about the task to be solved i.e. task-knowledge has to be supported. This kind of information enables the system to reason about its internal as well as about its environmental state in a (semi-)autonomous manner. This paper presents an approach for the logical verification of the input of such task-knowledge. The task-knowledge is represented on the basis of AND/OR-net structures. These net structures can be created easily by non-service robotic experts (application programmer e.g. care personnel). The logical verification during knowledge input on a semantic level becomes possible due to the enhanced a priori knowledge about the net elements (i.e. objects and operators) and their relationships.