Design of bicolored observers for interpreted Petri nets

This work presents an algorithm for the design of asymptotic observers for conservative discrete event systems under partial state observations modeled by interpreted Petri nets (IPN). The proposed observer consists of a Colored Petri net (CPN), where two token colors are defined: one for actual tokens and the other for estimated tokens. The observer uses the knowledge of the total amount of system resources and observed input-output sequences to set its initial marking, track the system evolution and rectify the marking estimation, resulting in a fast convergency.