Event-Triggered Real-Time Scheduling of Stabilizing Control Tasks
Summary (2 min read)
1. Introduction
- Small embedded microprocessors are quickly becoming an essential part of the most diverse applications.
- A particularly interesting example are physically distributed sensor/actuator networks responsible for collecting and processing information, and to react to this information through actuation.
- Common to all these approaches is the underlying principle that better control performance is achieved by providing more CPU time to control tasks.
- Close at the technical level, although addressing very different problems, is the recent work on stabilization under communication constraints [NE00, BL00, EM01, BPZ02, Lib03, NT04].
2. Notation and problem statement
- The authors shall not need the definition1 of ISS in this note but rather the following characterization.
- Definition 2.1. 1See, for example, [Son04] for an introduction to ISS and related notions.
- When ∆ > 0, the control task needs to be executed before the inequality γ(|e|) ≥ σα(|x|) is satisfied in order to account for the delay ∆ between measuring the state and updating the actuators.
- Answering the above questions is the objective of the following sections.
3. Existence of a lower bound for inter-execution times
- The authors start immediately with one of the main contributions of this note.
- Set R is forward invariant for the closed loop system since the execution rule (2.9) guarantees V̇ ≤ 0.
- Theorem 3.1 shows that the simple execution rule (2.9) results in a sequence of inter-execution times for the control task that is guaranteed to be lower bounded provided that ∆ is sufficiently small.
- The techniques used in the proof rely of Lipschitz continuity and are necessarily conservative for general nonlinear systems.
- For linear systems they provide reasonable estimates and one can even provide computable bounds for ∆ as discussed in the next section.the authors.
4. The linear case
- The authors also assume the existence of a linear feedback: u = Kx rendering the closed loop system globally asymptotically stable and where K is a matrix of appropriate dimensions.
- Note that in the linear case any such K renders the closed loop system ISS with respect to measurement errors.
- They are sufficiently accurate to be useful in practical situations as described in the next section.
5. An academic example
- In Figure 1 the authors can see how the error norm never reaches σ|x| even though it goes beyond σ′|x| which is used as execution rule.
- This gap decreases as the state approaches the origin.
- The authors can see that the estimated values, although conservative, do not overestimate the values obtained through simulation by more than a factor of 3.
6. Co-schedulability of stabilizing control tasks
- The authors shall assume a preemptive scheduler in which the control task has the highest priority and thus cannot be preempted by any other task and is executed without delays when γ(|e|) ≥ σα(|x|).
- Note that timing overheads associated with context switching can be captured in the proposed framework by suitably enlarging ∆.
- When a set of control tasks T can be scheduled despite the overhead associated with the control DR task the authors say that T is co-schedulable with the control task.
- Co-schedulability is now ensured by the following sufficient condition where dre denotes the smallest integer greater than r ∈ R. Theorem 6.1.
- Other possibilities are discussed in the next section.
7. Discussion
- 1. Real-time scheduling policies for non-preemptible tasks.
- The simple preemptive scheduling strategy presented in Section 6 relied on the possibility to preempt all but the control task.
- The results presented in this note are also relevant in this more general context since the lower bound on the inter-execution times can be used to construct a timed-automaton model for the control task.
- 2. ISS with respect to actuation errors and networked control systems.
- Similar ideas have been more or less explicitly explored in [YTS02, MA04, NT04].
Did you find this useful? Give us your feedback
Citations
48 citations
Cites methods from "Event-Triggered Real-Time Schedulin..."
...To cope with energy wasting in NCSs, the strategies of eventtriggered control (Tabuada, 2007; Heemels, Sandee and Van Den Bosch, 2008; Dimarogonas and Johansson, 2009; Wang and Lemmon, 2009b; Wang and Lemmon, 2008; Rabi, Johansson and Johansson, 2008; Henningsson, Johannensson and Cervin, 2008;…...
[...]
...To cope with energy wasting in NCSs, the strategies of eventtriggered control (Tabuada, 2007; Heemels, Sandee and Van Den Bosch, 2008; Dimarogonas and Johansson, 2009; Wang and Lemmon, 2009b; Wang and Lemmon, 2008; Rabi, Johansson and Johansson, 2008; Henningsson, Johannensson and Cervin, 2008; Henningsson and Cervin, 2010), and selftriggered control (Velasco, Marti and Fuertes, 2003; Wang and Lemmon, 2009a; Anta and Tabuada, 2010; Anta and Tabuada, 2009; Araujo, Anta, Mazo, Faria, Hernandez, Tabuada and Johansson, 2011; Mazo, Anta and Tabuada, 2010; Mazo, Anta and Tabuada, 2009; Millán Gata, Orihuela, Muños de la Peña, Vivas and Rubio, 2011) have been recently proposed....
[...]
48 citations
Cites background from "Event-Triggered Real-Time Schedulin..."
...Specifically, the event-triggering condition is usually constructed as the ratio of the state measurement error (which denotes the difference between the current signal and its latest sampled version) with respect to a predefined threshold function [24]....
[...]
...The events are generated by an event-triggering condition, which is a designed rule involving the state or output variables (see [24]–[26] and the references therein)....
[...]
47 citations
Additional excerpts
...t−d1 ?̇?(s)Q4?̇?(s)ds ≤ −[η(t) − η(t − d1)]Q4[η(t) − η(t − d1)], (31) and − (d2 − d1)∫ t−d1 t−d2 ?̇?(s)Q5?̇?(s)ds = − (d2 − d1)∫ t−d1...
[...]
47 citations
Additional excerpts
...…(Dimarogonas, Frazzoli, & Johansson, 2012; Guinaldo, Sánchez, & Dormido, 2016; Hu & Liu, 2016a; Hu, Liu, & Feng, 2016b; Mahmoud et al., 2016; Tabuada, 2007; Wang, Ni, & Ma, 2015; Xie, Xu, Chu, & Zou, 2015; Xie, Liao, & Li, 2016; Yin, Yue, & Su, 2014; Zhu, Guan, Luo, & Li, 2015; Zhang, Yang,…...
[...]
...In such a control scheme, the amount of communications and date update is reduced while the satisfactory performance is still remained by the reasonable triggered strategy, and a series of results about multi-agent systems event-triggered control have been reported (Dimarogonas, Frazzoli, & Johansson, 2012; Guinaldo, Sánchez, & Dormido, 2016; Hu & Liu, 2016a; Hu, Liu, & Feng, 2016b; Mahmoud et al., 2016; Tabuada, 2007; Wang, Ni, & Ma, 2015; Xie, Xu, Chu, & Zou, 2015; Xie, Liao, & Li, 2016; Yin, Yue, & Su, 2014; Zhu, Guan, Luo, & Li, 2015; Zhang, Yang, Yan, & Chen, 2015; Zhang, Yue, Yin, Hu, & Dou, 2016)....
[...]
47 citations
References
1,703 citations
"Event-Triggered Real-Time Schedulin..." refers background or methods in this paper
...This idea is an adaptation to the scheduling context of several techniques used to study problems of control under communication constraints [6], [11], [18]....
[...]
...Close at the technical level, although addressing very different problems, is the recent work on stabilization under communication constraints [6], [7], [11], [15], [18], [19]....
[...]
1,533 citations
"Event-Triggered Real-Time Schedulin..." refers background or methods in this paper
...This idea is an adaptation to the scheduling context of several techniques used to study problems of control under communication constraints [6], [11], [18]....
[...]
...Close at the technical level, although addressing very different problems, is the recent work on stabilization under communication constraints [6], [7], [11], [15], [18], [19]....
[...]
1,142 citations
"Event-Triggered Real-Time Schedulin..." refers background in this paper
...1See, for example, [24] for an introduction to ISS and related notions....
[...]
961 citations
799 citations
"Event-Triggered Real-Time Schedulin..." refers background in this paper
...Close at the technical level, although addressing very different problems, is the recent work on stabilization under communication constraints [6], [7], [11], [15], [18], [19]....
[...]