A Discretization Method from Coloured to Symmetric Nets: Application to an Industrial Example
Résumé
Intelligent Transport Systems (ITS) tend to be more distributed and embedded. In many cases, continuous physical parameters are part of the systems description. Analysis techniques based on discrete models must integrate such constraints. In this paper, we propose a methodological way to handle such hybrid systems with model checking on Petri Nets and algebraic methods. Our methodology is based on transformations from Coloured Petri Nets (CPN) for their expressiveness to Symmetric Petri Nets (SN) to take advantage of their efficient verification techniques. Our methodology also addresses the impact of discretization on the precision of verification. In scientific computing, the discretization process introduces “error intervals” that could turn a given verified property into a false positive one. In that case, the property to be verified might have to be transformed to cope with such precision errors.