No abstract

No abstract

International audienceThe software part of an automotive embedded system continues to increase significantly. It enables the development of new functionalities and it may improve the quality and comfort of driver assistance functions. However, the design of such functions becomes a complex task involving networked ECUs (Electronic Control Unit), several sensors/actuators and a set of embedded networks. The introduction of Model-Based Development (MBD) in the automotive field promised to improve the development process by allowing continuity between requirements definition, system design and the distributed system implementation. Further, the definition of AUTOSAR consortium standardized the design of such automotive embedded system by allowing the portability of software functions on the hardware architecture and their reuse. It defines a set of rules and interfaces to design, interconnect, deploy and configure a set of application software components (SWCs). However, designing an embedded system according to AUTOSAR standard necessitates the configuration of thousands of parameters and requires several software allocation decisions. Each decision may influence the system performance and also the development cost. This architectural complexity leads to a large design decision space which is difficult to be explored without using an analytical method or a design tool. For example, mapping software components (SWCs) to ECUs may affect the system performance. Actually, this phase of configuration and software allocation is performed manually using engineering and system architect knowledge. AUTOSAR provide a methodology for the software development of an Electricals/Electronics (E/E) system. However, this method doesn't guide the designer to deploy and bring a high-level software function onto a set of SWCs and then SWCs to ECUs. In this paper we present a model-based methodology to optimize software allocation and component configuration of an AUTOSAR system. This methodology relies on a multi-objective evolutionary algorithm which is characterized by its composability and speed performance. This algorithm is combined with a model-based system analysis engine permitting to evaluate system performance objectives. The objectives considered here are the CPU load, network load and functions response times

Abstract. In this paper we describe a technique for specifying time related properties on traditional software components. We apply the separation of concerns paradigm to allow independent specification of timing and to integrate timechecking specialized tool support into conventional software design processes. We aim at helping the designer to specify time contracts and at simplifying the introduction of time properties in the component behaviour description. We propose to handle timing issues in a separate and specific design activity, in order to provide means of formal computation of time properties for component assemblies without modifying in depth existing design processes. Scope and objectivesComponent based design is now at the heart of many modern applications. A rather important category of these applications must manage time, for instance because they interact with users in a time controlled manner (e.g. media players, group cooperation environments, etc) or because they are highly distributed (e.g. applications based on a bunch of Web services from diverse origins). Yet mainstream design techniques often emphasize type centric interactions between components: the component models they use offer powerful notations and tools for defining, refining and checking data types. Time properties are not explicitely taken into account by these models. At the source code level, programming languages and their associated frameworks also include some time characteristics [8]. Again, time propeties such as the maximum duration of an operation execution are treated as second class concepts: there are no time type systems. To overcome this deficiency, timeliness and other quality of service properties are sometimes specified using meta-attributes of programming languages (e.g. C# or Java). From a static validation point of view, these attributes are often treated like structured comments. These comments may be used to generate runtime monitors but their semantics is usually too weak to allow reasoning about time properties.At the design level, several research results have shown the usefulness of specific languages to describe component based software architectures. Thanks to the precise semantics of such languages, tools suites have been developed to analyze the consistency of a software architecture and to prototype it. For example, SOFA [17] provides a specific language that extends the OMG IDL to describe the architecture of component based software. It also provides a process algebra to specify the external behaviour This work was funded by ARTIST2, the Network of Excellence on Embedded Systems Design of component. However, using SOFA the architect cannot describe the required and provided QoS of components. The AADL standard [26] is one of the first ADL that provides mechanism to specify the QoS into the component interface, also identified as the fourth level of contract [7]. However, AADL is a low abstraction model, strongly connected with the implementation. Besides, AADL is not yet connected with tools tha...

In the domain of soft real-time application design, the gap between componentspecification models and the implementations often implies that the implementations cannot fully take advantage of the specification models. To limit this gap, this paper proposes an approach to generate a QoS monitor from the timed behavior specification. To support this approach, we rely on two different component models: one focused on formal description and the other on practical implementation. Those models are interconnected by model transformation, using a Model-Driven Engineering style.

In the domain of soft real-time application design, the gap between component-specification models and the implementations often implies that the implementations cannot fully take advantage of the specification models. To limit this gap, this paper proposes an approach to generate a QoS monitor from the timed behavior specification. To support this approach, we rely on two different component models: one focused on formal description and the other on practical implementation. Those models are interconnected by model transformation, using a Model-Driven Engineering style.

No abstract

Fast track article for IS&T International Symposium on Electronic Imaging 2021: Autonomous Vehicles and Machines proceedings.