Monitoring a hybrid system subjected to human actions needs both an estimation of the system state and an analysis of the possible conflicts in order to guarantee a safe behavior of the system. Such a monitoring principle is presented in this paper. The estimation of the hybrid system state is performed using the particle Petri net model. Then the estimated state is analysed by checking its consistency with respect to the reachable markings of the Petri net. The principle is applied to aircraft pilot’s activity tracking and conflict detection.