Search

International audience; The continuous integration of software-intensive systems together with the ever-increasing computing power offer a breeding ground for intelligent agents and multi-agent systems (MAS) more than ever before. Over the past two decades, a wide variety of languages, models, techniques and methodologies have been proposed to engineer agents and MAS. Despite this substantial body of knowledge and expertise, the systematic engineering of large-scale and open MAS still poses many challenges. Researchers and engineers still face fundamental questions regarding theories, architectures, languages, processes, and platforms for designing, implementing, running, maintaining, and evolving MAS. This paper reports on the results of the 6th International Workshop on Engineering Multi-Agent Systems (EMAS 2018, 14th-15th of July, 2018, Stockholm, Sweden), where participants discussed the issues above focusing on the state of affairs and the road ahead for researchers and engineers in this area.

International audience; To keep up with current technological developments, the engineering of multi-agent systems (MAS) has to provide solutions to: (i) support large scale systems, (ii) cope with open systems, and (iii) support humans in the loop. In this paper, we claim that the World Wide Web provides a suitable middleware for engineering MAS that address these challenges in an integrated manner. Even though approaches to engineer Web-based MAS have already been explored in the MAS community, existing proposals do not achieve a complete integration with the Web architecture. We approach this problem from a new angle: we design the agent environment as a hypermedia application. We apply REST, the architectural style of the Web, to introduce a resource-oriented abstraction layer for agent environments that decouples the application environment from its deployment context. Higher-level environment abstractions can then be implemented on top of this lower-level abstraction layer. To demonstrate our approach, we implemented a multi-agent application for the Internet of Things in which software agents can seamlessly navigate, use and cooperate in an environment deployed over multiple Web services (e.g., Facebook, Twitter) and constrained devices.

By abstracting devices to Web resources, the Web of Things (WoT) fosters innovation and rapid prototyping in the Internet of Things (IoT): it enables developers to use standard Web technologies for creating mashups of Web services that perceive and act on the physical world (a.k.a. physical mashups). In recent years, however, it has become apparent that current programming paradigms for Web development have important shortcomings when it comes to engineering IoT systems: static Web mashups cannot adapt to dynamic IoT environments, and manually mashing-up the IoT does not scale. To address these limitations, WoT researchers started to look for means to engineer WoT systems that are more autonomous in pursuit of their design objectives. The engineering of autonomous systems has already been explored to a large extent in the scientific literature on artificial intelligence. In this position paper, we distill that large body of research into a coherent set of abstractions for engineering autonomous WoT systems.

To cope with dynamic environments, Internet of Things (IoT) applications are expected to autonomously discover and interact with services at runtime in pursuit of design or user-specified goals. On the one hand, various paradigms and technologies are available to program goal-driven autonomous software agents, and on the other hand hypermedia-driven environments are central to the development of robust machine-to-machine applications. However, existing approaches for the development of hypermedia-driven environments fall short of meeting the needs of autonomous agents: they either severely restrict the agents' autonomy, or their topological structure is either fragmented or inefficient to navigate at scale. In this paper, we explore the use of socio-technical networks, that is networks of people and things interrelated in a meaningful manner via typed relations, as an overlay for enhancing hypermedia-driven interaction in IoT environments. We present a proof of concept and discuss several classes of applications in which this model could prove useful.

International audience; Applications envisioned for the Internet of Things (IoT) would generally have to fulfill their design goals by mashing up devices and digital services in a manner that is both flexible, such that they can adapt to dynamic environments, and responsive, such that they can react to sensor and user input in a timely fashion. Most existing approaches for the development of IoT applications rely on precompiled mashups that are highly responsive, but inflexible due to their static nature. At the other end of the spectrum, fully automatic composition of services results in IoT mashups that are highly flexible, but responsive only for small numbers of IoT services. This paper presents a middle ground approach: goal-driven software agents are equipped with precompiled mashups and cooperate with one another to compose their mashups at runtime in pursuit of their goals. Agents are interconnected via relations that enable them to discover and interact with one another in a flexible manner. To support our approach, we provide an open-source platform that facilitates application development. We used this platform to implement a realistic IoT application that achieves its design goal by mashing-up multiple heterogeneous devices at runtime. Evaluation results suggest that applications remain responsive when scaling to many devices and for relatively large mashup compositions.

The Web is to experience considerable growth by extending to physical devices, i.e. the so-called Web of Things (WoT). To make this growth sustainable, users need mechanisms for managing and interacting with large networks of devices and services. Existing search engines and mashup editors for the WoT partially address this problem by enabling developers and tech savvy users to search for "things" and connect them to one another. However, manually "wiring" large networks of things does not scale. Furthermore, static mashups cannot adapt to dynamic environments. Our approach is to apply the social network metaphor to theWoT to create a semantic socio-technical overlay that may be reliably processed and used by both people and things. Things can crawl this overlay to discover one another, but also manipulate it such that they may wire themselves. In this paper, we introduce a model for socio-technical networks (STNs). We apply this model to wrap up Twitter and Facebook as STNs, and to develop ThingsNet, our own prototype STN platform. Finally, we present a software client that uses the STN model to function across all three platforms, as well as any other platforms that may be translated to this model.

Developing applications across the physical-digital space requires the homogeneous interconnection of people, physical devices, services and various data sources as first-class entities of complex socio-technical systems. In this paper, we describe socio-technical networks (STNs) as the building blocks of a semantic, open and distributed Social Web of Things. We address the problem of enabling autonomous non-human agents as participants in an open set of STNs. Our approach is to provide agents with machine-readable descriptions of STNs, of operations required for participating in such systems, and of supported implementations for those operations. Towards this aim, we present the STN ontology and we illustrate its applicability. Even though the STN ontology is a work in progress, using the core concepts and properties described in this paper we are able to create concrete specifications of STN platforms. We discuss the positioning of this ontology with respect to several well-known and related vocabularies.

International audience; The notion of a Social Web of Things (SWoT) appears in recent works at the convergence of the Social Web and the Web of Things. In our vision, a third dimension is needed: pro-activeness. We propose to extend and transform social networks by integrating autonomous and proactive things. In this paper, we discuss the evolution of the Web on several dimensions, leading to our vision for the SWoT. We discuss the challenges that need to be addressed, a possible approach for addressing them and we illustrate the applicability of the SWoT through a motivating scenario.

People do not generally mind that some personal information is shared, but that it is shared in the wrong ways and with inappropriate others. We introduce a model for the automatic detection of incompatible relationships among agents in a virtual community. We define this model to the aim of adding to the PrivaCIAS framework a new layer that follows the theory of contextual integrity in more depth. We describe in detail the implementation of a proof of concept privacy-aware photo-sharing social network that we had developed for Android smartphones. The social network was built on the extended PrivaCIAS framework, and thus illustrates its applicability in helping users to preserve privacy.