Design of robust robot programs: Deviation detection and classification using entity-based resources

Robot applications are located more and more often in highly dynamic environments. In contrast to well known, structured environments, this leads to the challenging problem of keeping the robot programs robust. One aspect of robustness is the ability to detect and handle unexpected events, e.g. a dropped object or a storage place, which should be free, but is already occupied. In this work we call such events deviations. The contribution of this paper is to show, that we can design robot programs more robust, when we regard these deviations. We propose an approach that provides the detection and classification of deviations occurring during the execution of a robot program. Furthermore we show that the classified deviations can be used to develop a customized deviation management. For this purpose, an Entity-Component-System (ECS) is used to describe any relevant resources in the workspace of the robot. With such a resource model we are able to detect whether there is a difference between the expected state and the actual state of the relevant robot environment. Based on that monitoring our approach provides a statement about the presence and type of a deviation. The advantages of these approach including a unified detection and classification principle and a base for recovering from failures.