A TAOPN Approach to Modeling and Scheduling Cyber-Physical Systems

Cyber-physical system is a combination of computation, control and communication. Existing modeling approaches are not suitable for the implementation of its complicated behaviors especially the mixing of discrete and continuous parts. Due to this feature, we propose a time-constrained aspect-oriented Petri net (TAOPN) approach to model cyber-physical system. This approach extends Petri net with time constraints based on aspect-oriented method. Common physical processes are extracted out and modeled as continuous Petri net in an aspect and then woven into the basic discrete Petri net according to the weaving mechanism. Those physical processes presented in an aspect net can be reused and analyzed by continuous analysis tools without interfering with the discrete model. The whole system can still be analyzed and verified using formal methods. A general scheduling analysis method is presented. This method decomposes the model into four types of modules and analyzes their schedulability separately. Then composes these results into a whole schedulable sequence. A case study demonstrates this modeling approach is feasible and the schedulability analysis is applicable.