Scheduling packets for Event-triggered control

For a single control loop with communication rate constraints, Event-triggered control often outperforms periodic control. When multiple loops are being controlled over a shared contention-based medium, however, the advantage of event-triggered policies is less well understood. In this paper, we consider event-triggered impulse control under lossy communication. The sampling events are determined by level crossings of the plant output. It is shown how a stochastic control criterion depends on the level thresholds and the packet loss probability for a class of integrator plants. For multiple control loops, this result is used to derive a design guideline on how to assign the levels that lead to optimal use of the available communication resources. It is shown that the structure of the event generator depends critically on the loss probability.

[1]  Ioannis Karatzas,et al.  Brownian Motion and Stochastic Calculus , 1987 .

[2]  Anton Cervin,et al.  Sporadic Control of First-Order Linear Stochastic Systems , 2007, HSCC.

[3]  Maben Rabi,et al.  Packet based Inference and Control , 2006 .

[4]  K. Åström,et al.  Comparison of Riemann and Lebesgue sampling for first order stochastic systems , 2002, Proceedings of the 41st IEEE Conference on Decision and Control, 2002..

[5]  Anton Cervin,et al.  Scheduling of event-triggered controllers on a shared network , 2008, 2008 47th IEEE Conference on Decision and Control.

[6]  Hermann Kopetz,et al.  Real-time systems , 2018, CSC '73.

[7]  Karl Henrik Johansson,et al.  Optimal stopping for event-triggered sensing and actuation , 2008, 2008 47th IEEE Conference on Decision and Control.

[8]  Ronald W. Wolff,et al.  Poisson Arrivals See Time Averages , 1982, Oper. Res..

[9]  Mikael Johansson,et al.  Networked estimation under contention-based medium access , 2010 .

[10]  J.S. Baras,et al.  Multiple Sampling for Estimation on a Finite Horizon , 2006, Proceedings of the 45th IEEE Conference on Decision and Control.