Health Monitoring of Hybrid Systems Using Hybrid Particle Petri Nets

This paper presents an approach of model-based diagnosis for the health monitoring of hybrid systems. These systems have both continuous and discrete dynamics. Modified Parti-cle Petri Nets, initially defined in the context of hybrid sys-tems mission monitoring, are extended to estimate the health state of hybrid systems. This formalism takes into account both uncertainties about the system knowledge and about di-agnosis results. The generation of a diagnoser is proposed to track online the system health state under uncertainties by us-ing particle filter. To include more complex characteristics of the system, as its degradations for prognosis purpose, an en-riched formalism called Hybrid Particle Petri Nets is defined. 1