On the Duality Between Event-Driven and Time-Driven Models

2016

Christos G. Cassandras, without whom I would have never been able to be where I am right now. His sincere interest in my success, his patience, motivation, and incredible amount of knowledge have been the light on my way. His guidance helped me find my path in research in many ways. I could not have imagined having a better advisor and mentor for my PhD study. Besides my advisor, I would like to sincerely thank the rest of my dissertation committee, Prof. Paschalidis, Prof. Vakili and Prof. Andersson. Thanks for your cooperation, guidance and most of all your ideas and questions. I also thank Prof. Caramanis for accepting to serve as the chair for my PhD defense. During my PhD journey, everybody in the Division of Systems Engineering have been kind, respectful, and always available for help. I would like to specially thank

Proceedings 1995 INRIA/IEEE Symposium on Emerging Technologies and Factory Automation. ETFA'95, 1995

The automation of batch processes poses clifficult issues because it is necessary to concurrently deal with continuous and discrete models. In this paper we present a Petri net based approach which differs from that of [7, 121 because we do not employ a unified framework based on extended Petri nets. Indeed, we do not try to integrate the continuous aspect within the fmmework of Petri net theory. In contrast, we establish a cooperation and powerful interaction between the Petri net model of the discrete aspect of the batch system and the continuous model which is a set of differential algebraic equations. This approach has been tested in the food industry for the validation of scheduling policies and is currently been developed for supervisory control and reactive scheduling.

2000

In software developments, the simple succession of operations can easily be hard-coded. However, to control event driven systems, applications must conform to a well-defined sequence of operations, even simultaneous ones, that depends on a variable environment, on time constraints and on synchronization constraints between several sequences. In that kind of applications, too complex intertwined conditional statements must be avoided as

2011

The paper present and compare some real-time control program models used by programmers programming in embedded systems. The paper do not demonstrate all programming models, only the most important variations based on writer's opinion. The writer tried to show the advantages and disadvantages of each programming models discussed in article. The paper begins with traditional models like sequential model, cooperative multitasking and continues with timed cooperative model and finish with event-driven model.

IEEE Transactions on Automatic Control, 1994

This paper extends the authors' previous work on observahillity of discrete-event systems by taking time into consideration. In a timed discrete-event system, events must occur within their respective lower and upper time bounds. A supervisor can disable, enable, or force some events to achieve a given control objective. We assume that the supervisor does not observe all events, which is often the case in practice. We generalize the concept of observability to timed discrete-event systems and show that it characterizes the existence condition for a supervisor. We also generalize normality, a stronger version of observability, to timed discrete-event systems, which has nice properties that are absent in observability. We then derive conditions under which observability and normality are equivalent. We propose two methods to synthesize a supervisor, a direct approach and an indirect approach. An example is given to illustrate the results.

In supervisory control theory, an autonomous plant and its specification are modelled as sets of event sequences. The task of a supervisor is to control the plant by disabling certain events so that the obtained behaviour lies within the specification. We extend this theory to the case where the plant and the specification contain temporal constraints and are described by a model called Timed Automata (TA). The problem that arises with TA is that the state space can be infinite due to the infinite number of time values. In a recent work, we proposed an efficient method to finitely represent the (infinite) state space. The proposed method transforms a TA describing a timed system into a minimal and equivalent finite state automaton (FSA) using two special types of events : Set and Exp. Such a FSA is denoted se-FSA. In this article, we propose a method for the supervisory control of timed discrete event systems that are modelled by TA. First, we use the transformation procedure for representing the plant and the specification as two se-FSAs. And second, we develop a procedure for generating the supervisor from the two se-FSAs that represent the plant and the specification. We also propose a simple supervisory control architecture.

Lecture Notes in Computer Science, 1996

Event structures are a promi nent noni nterl eavi ng model for concurrency. Real-ti me event structures associ ate a set of ti me i nstants to events, model l i ng absol ute ti me constrai nts, and to causal dependenci es,model l i ng rel ati ve del ays betweencausal l y dependentevents. We i ntroduce thi s novel temporal model and show how i t can be used to provi de a denotati onal semanti cs to a real-ti me vari ant of a process al gebra aki n to LOTOS. Thi s formal i sm i ncl udes a ti med-acti on pre x whi ch constrai ns the occurrence ti me of acti ons, a ti meout and watchdog (i .e., ti med i nterrupt) operator. An event-based operati onal semanti cs for thi s formal i sm i s presented that i s shown to be consi stent wi th the denotati onal semanti cs. As an exampl e we use an i n ni te bu er wi th ti me constrai nts on the message l atency and the rates of accepti ng and produci ng data. ? The work in this paper is partially funded by C. N. R.-Progetto Bilaterale: Estensioni probabilistichee temporali dell' algebra di processi LOTOS basate su strutture di eventi, per la speci ca e analisi quantitative di sistemi distribuiti, by C. N. R.-Progetto Coordinato: Strumenti per la speci ca e veri ca di proprieta' critiche di sistemi concorrenti e distribuiti, and by the EU as part of the ESPRIT BRA project 6021: Building Correct Reactive Systems (REACT).

IEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans, 2001

This paper is an extension of an earlier work on a methodology for modeling temporal aspects of discrete-event systems. The methodology incorporates point and interval descriptions of time, and offers both qualitative and quantitative calculus for time. A graph-based temporal programmer (TEMPER) is shown to implement the axiomatic system of the temporal formalism. The approach transforms the system specifications given by temporal statements into a graph structure, identifies errors (if present) in the system, infers new temporal relations among system intervals, and calculates delays among time points and their actual time of occurrence.

18th Mediterranean Conference on Control and Automation, MED'10, 2010

A method has been recently proposed for the supervisory control of real-time discrete event systems (RTDES). This method is based on a transformation, called SetExp, of timed automata (TA) into finite state automata. This SetExpbased method is computationally less complex than other realtime control methods, but it may synthesize less permissive supervisors. This paper improves significantly the SetExpbased control method: 1) we improve the permissiveness of the supervisors; 2) we study and present more rigorously and completely the control method and its architecture.

Lecture Notes in Computer Science, 2004

The design of reactive systems must comply with logical correctness (the system does what it is supposed to do) and timeliness (the system has to satisfy a set of temporal constraints) criteria. In this paper, we propose a global approach for the design of adaptive reactive systems, i.e., systems that dynamically adapt their architecture depending on the context. We use the timed automata formalism for the design of the agents' behavior. This allows evaluating beforehand the properties of the system (regarding logical correctness and timeliness), thanks to model-checking and simulation techniques. This model is enhanced with tools that we developed for the automatic generation of code, allowing to produce very quickly a running multi-agent prototype satisfying the properties of the model.