Abstract We present , a framework for the integration of modern verification tools in the context of AUTOSAR, a widely-used open standard for the development of automotive software systems. Our framework enables the automatic end-to-end verification of system-level properties using a compositional approach. It combines software model checking techniques for the verification of software components at the code level with a contract-based analysis for verifying their correct composition. In this paper, we present the tool through its application on a representative automotive case study, discussing the main functionalities provided and the results obtained.

Analysis models, technologies and tools are extensively used in the automotive domain to validate and optimize the design and implementation of SW systems. This is especially true for modern systems including advanced autonomous (and complex) features. The range of analysis methods that can be applied is extremely wide and goes from functional correctness to functional safety to timing (and schedulability), security, and possibly even more. The AUTOSAR automotive standard has been defined with the purpose of standardizing the SW architecture of automotive systems and enable the construction of systems by composing SW components that are portable and abstract with respect to the underlying HW/SW platform. However, AUTOSAR was originally developed with portability of code in mind, and even if it quickly evolved to include a system-level modeling language (with its metamodel) and later extensions to deal with the needs of analysis methods (and tools), it is hardly comprehensive and still affected by several omissions and limitations. To fix the limitations with respect to timing and schedulability analysis Bosch developed the Amalthea (later App4MC) metamodel and tools. In Huawei, a more general (and ambitious) approach was undertaken to support not only timing analysis, but also model checking (or other types of formal verification), safety analysis and even design optimization. The approach is based on the concepts of a unified (modular) metamodel and a framework based on Eclipse to integrate analysis methods and tools. In this paper we describe the framework and the results obtained with respect to the objectives of functional verification and timing analysis.

Web services represent a promising technology for the development of distributed heterogeneous software systems. In this setting, a major issue is to establish whether two services can be used interchangeably in any context. To this aim, our paper first briefly reviews the results contained in a recent article by the same authors, where a suitable notion of behavioural equivalence for Web services was introduced. Our work then extends those results, in order to account for ontologybased service specifications. Next, a concrete example scenario – a car rental system – is presented, and it is then used to illustrate how the equivalence between services can be fruitfully employed for correctly addressing two prominent, modularity-related problems: the publication of correct service specifications and the replaceability of (sub)services.

We present a service discovery architecture for embedded peer-to-peer systems, tailored to deal with low-capacity, mobile devices. The proposed discovery mechanism is built on top of the (Chord) Distributed Hash Table technology, extending it by fruitfully exploiting high-capacity devices (when avai

Web service discovery is one of the key issues in the emerging area of Service-oriented Computing. In this paper, we present a complete composition-oriented, ontology-based methodology for discovering semantic Web services, which exploits functional and behavioral properties contained in OWL-S service advertisements to satisfy functional and behavioral client queries. To this aim, we build on top of the results contained in two recent articles, where we presented (1) a suitable data structure (viz. a dependency hypergraph) to collect functional information of services, and (2) a suitable notion of behavioral equivalence for Web services. We also discuss the architecture and the main implementation choices of the matchmaking system applying such a methodology.

Service discovery and service aggregation are two crucial issues in the emerging area of service-oriented computing (SOC). We propose a new technique for the discovery of (Web) services that accounts for the need of composing several services to satisfy a client query. The proposed algorithm makes use of OWL-S ontologies, and explicitly returns the sequence of atomic process invocations that the client must perform in order to achieve the desired result. When no full match is possible, the algorithm features a flexible matching by returning partial matches and by suggesting additional inputs that would produce a full match.