Developing Control Systems with Some Fragile Environment?

mail: Jean-Luc.Paillet@cmi.unv-mrs.fr ABSTRACT In this paper, we present an original discrete event formalism for the specification of control systems at a high level of abstraction. The concept of event being more natural than the concept of state, the proposed formalism is centered on this concept of event. It allows the expression of an asynchronous behaviour using this concept. In addition, this formalism offers an explicit representation of time that allows to use temporal simulations for the validation of the formal specifications, after translation into DEVS models.

Procedia Computer Science, 2011

Today, distributed systems with loose connections are increasingly developed using event-based architectures. In these architectures, systems are composed of separate components which cooperate dynamically. The communication and cooperation of components in these architectures are done through event passing mechanisms. The main challenge in these architectures is to define a framework to model and validate systems under development. In this paper, we propose a framework using ASMETA (ASM mETA language) for modelling systems through event-based architectures. Validation of the whole system is done through model checking using Bogor. We describe the process of automatically mapping ASM models to BIR -the input language of Bogor-. In the end, we present our experimental results using the proposed approach with different case studies.

Computer Science and Information Systems, 2020

Embedded control systems combine digital and physical components, leading to complex interactions and even complexity of their development. In [4] we proposed a method to build such complex systems in a systematic way. The overall method starts from an abstract model of the physical environment of the considered system and its controller. The method consists in a sequence of refinement steps, in the spirit of Event-B, that gradually introduces design details from an abstract level, until more concrete levels. Two main refinement processes are distinguished: one to capture the global model, the other to detail it; we provide through the method the guidelines to accompany these two refinement processes. But there were a lack of assistance tools. The designers need to be assisted by tools to guide them, to automate partially the refinements and to help in proving more easily model properties. We illustrate the method with the landing gear system case study and choosing the Event-B tool...

2018

We propose a method to build critical embedded control systems in a systematic way. The method covers the modelling of both the digital part and the physical environment of a considered system, and their refinement until more concrete levels. It is based on Event-B in order to benefit from its materials, stepwise refinements and tools. Two main processes are distinguished: one to capture the global model, the other to detail the global model; they are made of several refinement steps which are accompanied with guidelines. The precise description of the interface between the digital and physical parts is used to start the modelling process. The recurrent categories of variables and events in control systems are described and used as guidelines to conduct a systematic construction. We illustrate the method with the landing gear system case study.

Lecture Notes in Computer Science, 2014

Evolving distributed systems are made of several physical devices distributed through a network and a set of functionalities or applications hosted by the physical devices. The configuration of the physical components may be modified through the time, hence the continuous evolving of the whole system. This should affect neither the hosted software components nor the global functionning of the whole system. The components of the systems are software components or physical components but their abstract models are considered with the aim of modelling and reasoning. We show that an event-based approach can be benefically used to model and verify this kind of evolving control systems. The proposed approach is first presented, then the CCTV case study is introduced and modelled. The resulting model is structured as a B abstract machine. The functional properties of the case study are captured, modelled and proved. The refinement technique of Event-B is used to introduce and prove some properties.

Lecture Notes in Computer Science, 2003

The event based architectural style has been recognized as fostering the development of large-scale and complex systems by loosely coupling their components. It is therefore increasingly deployed in various environments such as middleware for mobile computing, message oriented middleware, integration frameworks, communication standards, and commercial toolkits. The development of applications based on this paradigm is, however, performed in such an ad-hoc manner that it is often difficult to reason about their correctness. This is partly due to the lack of suitable specification and verification techniques. In this paper, we review the existing theory of specifying and verifying such applications, argue that it cannot be applied for the development of large-scale and complex systems, and finally propose a novel approach (LECAP) for the construction of correct event based applications. Our approach is superior to the existing approaches in many respects: 1) we assume a while-parallel language with a synchronization construct, 2) neither a pending event infrastructure nor a consume statement are required, 3) a dynamic (instead of static) binding is assumed, 4) no restriction is made on the number of simultaneous executions of the same program 5) our approach is oriented towards top-down development of systems. The paper also presents two examples for illustrating the approach.

SMC'98 Conference Proceedings. 1998 IEEE International Conference on Systems, Man, and Cybernetics (Cat. No.98CH36218), 1998

In this paper, we present the integration of a controller synthesis methodology in the Signal environment through the description of a tool dedicated to the algebraic computation of a controller and then to the simulation of the controlled system. The same language is used to specify the physical model of the system and the control objectives. The controller is then synthesized using the formal calculus tool Sigali. The result is then automatically integrated in a new Signal program in order to obtain a simulation of the result.

22nd International Symposium on Reliable Distributed Systems, 2003. Proceedings., 2003

Abstract Today's software technology was created for applications that used a relatively small num ber of homogeneous devices Current applications need to deal with thousands, and in the future perhaps millions, of highly heterogeneous devices research in distributed systems is currently trying to invent software methodologies con sisting of communication paradigms, tools, mechanisms, and techniques that overcome the limitations of current languages and techniques purpose is the event - based communication paradigm (also called implicit invocation or publish/subscribe) The primary benefit of the event - based paradigm is that it supports the loose coupling of components that compose an application and therefore scales to large numbers of hetero geneous components The use of the event - based architectural style has been successfully demonstrated in the development of large - scale and complex systems been rapidly incorporated in not only research prototypes but also commercial products and toolkits and even in software communication standards development based on this paradigm is, however, ad hoc and informal often di cult to reason about the correctness of the resulting applications systematic and rigorous basis for the development of event serious problem as the event - based paradigm is being used increasingly in important do mains such as flight - control, e - commerce, automotive, and home applications theory of specifying and verifying such applications cannot be applied for the development of large - scale and complex systems

Computer Languages, 1991

Abstraet--This paper presents PROTOB, an object oriented language and methodology based on PROT nets, and the CASE environment that supports it. PROT nets integrate extended dataflows and Petri nets into an object oriented formalism. The CASE environment consists of several tools supporting specification, modelling and prototyping activities using the PROTOB language within the framework of the operational software life cycle paradigm. As its major application area it addresses distributed systems, such as real-time embedded systems, communication protocols and manufacturing control systems. The CASE environment automatically generates the distributed and object oriented implementation code in Ada or C providing advanced features such as multitasking and system distribution over a LAN. A model of a Flexible Manufacturing System and its production control software are analysed as a case study. Automatic code generation Dataflows Distributed systems Executable specification Object oriented design Object-oriented specification Operational software life cycle Petri nets PROT nets Rapid prototyping Structured analysis 2 Send_Id DEVICE.DEVICE.TOPVIEW Ready, Nui © P~eply.Mess&se Request.Message Serve . Device object.

2004

Various application domains exist where the advantages of the event-based paradigm make it a key technology. In general, this architectural style allows better control of the structural and behavioral complexity of applications: com- ponents can be developed independently and loosely inte- grated. The computational behavior of this paradigm, how- ever, remains poorly understood. This position paper ar- gues on the necessity of a new methodology for constructing event-based applications as well as a new logic that clari- fies the computational behavior of such applications. For this, the paper presents some factors that make the event- based style so troublesome and discusses the (non-) ade- quacy of existing formal techniques for the construction of event-based applications.