Francesco L. De Angelis

Developing pervasive multi-agent systems with nature-inspired coordination

Pervasive computing systems can be modelled effectively as populations of interacting autonomous components. The key challenge to realizing such models is in getting separately-specified and -developed sub-systems to discover and interoperate with each other in an open and extensible way, supported by appropriate middleware services. In this paper, we argue that nature-inspired coordination models offer a promising way of addressing this challenge. We first frame the various dimensions along which nature-inspired coordination models can be defined, and survey the most relevant proposals in the area. We describe the nature-inspired coordination model developed within the SAPERE project as a synthesis of existing approaches, and show how it can effectively support the multifold requirements of modern and emerging pervasive services. We conclude by identifying what we think are the open research challenges in this area, and identify some research directions that we believe are promising.

Logic Fragments: A Coordination Model Based on Logic Inference

Chemical-based coordination models have proven useful to engineer self-organising and self-adaptive systems. Formal assessment of emergent global behaviours in self-organising systems is still an issue, most of the time emergent properties are being analysed through extensive simulations. This paper aims at integrating logic programs into a chemical-based coordination model in order to engineer self-organising systems as well as assess their emergent properties. Our model is generic and accommodates various logics. By tuning the internal logic language we can tackle and solve coordination problems in a rigorous way, without renouncing to important engineering properties such as compactness, modularity and reusability of code. This paper discusses our logic-based coordination model and shows how to engineer and verify a simple pattern detection example and a gradient-chemotaxis example.

Logic Fragments: Coordinating Entities with Logic Programs

Rigorous engineering of self-organising and self-adaptive systems is a challenging activity. Interactions with humans and unexpected entities, dependence on contextual information for self-organisation and adaptation represent just some of the factors complicating the coordination process among multiple entities of the system. Recently we proposed a coordination model based on logic inference named Logic Fragments Coordination Model. Logic Fragments are combinations of logic programs defining interactions among agents distributed over the nodes of the system. They are able to accommodate various types of logics, ranging from classical up to many-valued paraconsistent ones. The logical formalisation makes it possible to express coordination in a rigorous and predicle way, both at design-time and run-time. In this paper we define, under the form of an evaluation algorithm, the semantics of Logic Fragments; introducing logical predicates used to manage and reason on local and remote information. By associating specific semantics to the symbols inferred during the evaluation of logic programs it is possible to make logical inference effects unambiguous on the system; such an approach turns Logic Fragments into a coordination-oriented logic-based programming model. We conclude the paper discussing three examples showing the use of Logic Fragments to implement on-the-fly ad-hoc coordination mechanisms, as well as design-time and run-time verification of spatial properties.

A Logic Language for Run Time Assessment of Spatial Properties in Self-Organizing Systems

The assessment of emergent global behaviors of self-organizing applications is an important task to accomplish before employing such systems in real scenarios, yet their intrinsic complexity make this activity still challenging. In this paper we present a logic language used to verify graph-based global properties of self-organizing systems at run-time. The logic language extends a chemical-based coordination model based on logic inference recently proposed. The logic formulae defined by using the language operators depict the intended global spatial properties arising from local interactions among components. Logic formulae are evaluated in a distributed manner by using an inference procedure which checks them against the current global state of the system, verifying whether the intended emergent global behavior actually appears in the system. As examples of spatial properties we consider color patterns: at first we show how to verify specified patterns of identified colors in sets of nodes directly connected, then we present other formulae verifying the appearance of global patterns of colors without specifying the colors themselves. We conclude the examples with the computation of mathematical functions, like the verification of the existence of a maximum value in a specific node of the system.

The ONE-SAPERE Simulator: A Prototyping Tool for Engineering Self-Organisation in Pervasive Environments

This paper presents The ONE-SAPERE simulator, the first simulator combining an opportunistic network environment simulator with a middleware for pervasive systems, the SAPERE Middleware, which has already been released for Android devices and PCs.

Secure Channel Service for MANETs

This paper shows how to use a well defined set of self-organizing services for establishing and preserving a confidential and adaptive channel between two communicating entities on top of an ad-hoc mobile network.

Spatial edge services

Ubiquitous and context-aware sensors are increasing in number and aim at providing comfort and better life quality. They are spatially distributed and their computation capacity are still under-exploited. This article discusses spatial edge services, a new generation of services exploiting IoT and spatially distributed data. They result from collective and decentralized interactions of multiple computing entities. They rely on a logic and chemical-based coordination model. In this paper, we discuss several spatial edge services cases and present an actual deployment and a prototype example. We discuss evaluation results, in particular scalability and accuracy. Spatial edge services provide innovation capabilities for the software industry, connected objects manufacturers and edge computing industry. Future work tackles on-the-fly service composition with requested quality of service (QoS).

Heterogeneous Approximate Reasoning with Graded Truth Values.

This paper is devoted to paraconsistent approximate reasoning with graded truth-values. In the previous research we introduced a family of many-valued logics parameterized by a variable number of truth/falsity grades together with a corresponding family of rule languages with tractable query evaluation. Such grades are shown here to be a natural qualitative counterpart of quantitative measures used in various forms of approximate reasoning. The developed methodology allows one to obtain a framework unifying heterogeneous reasoning techniques, providing also the logical machinery to resolve partial and incoherent information that may arise after unification. Finally, we show the introduced framework in action, emphasizing its expressiveness in handling heterogeneous approximate reasoning in realistic scenarios.

Towards a Spatial Language for Run-Time Assessments in Self-Organizing Systems

In this paper we define a spatial language used to verify global properties of self-organizing systems at run-time. The language can be used to assess spatial properties of system components to check desired global properties of the system against emergent global behaviors arising from local interactions among components. The spatial language extends a logic-chemical-based coordination model that we have recently proposed and the verification of spatial properties is performed in a distributed manner among the nodes of the system.

Self-composition of services with chemical reactions

Next generation of socio-technical infrastructures will be characterized by the presence of complex networks of pervasive systems, composed of thousands of heterogeneous devices consuming and producing high-volumes of interdependent data. Smart-cities represent an example of these future digital scenarios: by using wide area mobile ad-hoc networks (MANETs), data will be shared among applications placed on cars or running on several devices such as smartphones, tablets, public displays and sensors placed at the edges of the roads; moreover, all these devices will access traditional remote web-services. Smart-cities depict the emergence of new open-infrastructure pervasive systems, where scalability and dependability will be achieved by developing and adapting (at run-time) applications through compositions of customized services.