Event-triggered dual-mode distributed model predictive control

Xu SU,Yuan-yuan ZOU,Yu-gang NIU,Ting-gang JIA
DOI: https://doi.org/10.7641/CTA.2016.50964
2016-01-01
Abstract:This paper proposes an event-triggered dual-mode distributed model predictive control method for large-scale linear discrete-time systems subject to bounded disturbances. The event-triggering condition, which involves information of the subsystem itself, is obtained by the input-to-state stability (ISS) theory. Only when the event-triggering condition is satisfied, the state measurement is sent and the distributed model predictive control optimization problem is solved. Meanwhile, the subsystem exchanges its optimal state trajectories with neighbor subsystems. When the subsystem enters the invariant set, the state feedback control law will be applied. Moreover, no information will be exchanged between the subsystem and its neighbor subsystems which also enter the invariant sets. The upper bound of disturbances are derived by analyzing the recursive feasibility and closed-loop stability. Finally, the algorithm is verified through the vehicle control systems. Simulation results show that the presented method is able to reduce the solving frequency of optimization problems and the number of information transmissions, thus saving the computation resources and communication resources.
What problem does this paper attempt to address?