Sensor fusion: a rough Petri net model

This paper introduces the use of rough Petri nets to model sensor fusion. Considerable work has already been carried out in modeling various forms of systems with Petri nets. The aim of the earlier as well as the current research has been to provide a complete framework for approximate reasoning. This paper returns to the idea of rough Petri nets to model sensor fusion. Guarded transitions based on multivalued logic and rough sets are used to design Petri net models of sensors and filters. Dill receptor processes are used to define input places that model sensors. Rough integrals are used to design particular forms of Lukasiewicz guards. The contribution of this paper is the modeling of fusion of sensors considered relevant in a problem-solving system. An application of this approach is given in fusing homogenous sensors