Multi-Robot System (MRS) is composed of a group of robots that work cooperatively. However, Multi-Agent System (MAS) is computational systems consisting of a group of agents that interact with each other to solve a problem. The central difference between MRS and MAS is that in the first case, the agent is a robot, and in the second, it is a software. Analyzing the scientific literature, it is possible to notice that few studies address the integration between MAS and MRS. In order to achieve the interdisciplinary integration, the theoretical background of these areas must be considered in this paper, so that the integration can be applied using a case study of decentralized MRS. The objective of this MRS is to track and surround a stationary target. Also, it has been implemented and validated in the robot simulator called Virtual Robot Experimentation Platform (V-REP). In the validation of the proposed MRS, a scenario with three robots and a stationary target were defined. In the tracking task, the robot can detect the target whose position is not known a priori. When the detection occurs, the V-REP informs the target position to the robot because the environment is discretized into a grid of rectangular cells. After that, all the robots are directed to the target, and the surround task is realized. In this task, a mathematical model with direct communication between the robots was used to keep the robots equidistant therefrom and from each other.

Selecting pivot features that connect a source domain to a target domain is an important first step in unsupervised domain adaptation (UDA). Although different strategies such as the frequency of a feature in a domain, mutual (or pointwise mutual) information have been proposed in prior work in domain adaptation (DA) for selecting pivots, a comparative study into (a) how the pivots selected using existing strategies differ, and (b) how the pivot selection strategy affects the performance of a target DA task remain unknown. In this paper, we perform a comparative study covering different strategies that use both labelled (available for the source domain only) as well as unlabelled (available for both the source and target domains) data for selecting pivots for UDA. Our experiments show that in most cases pivot selection strategies that use labelled data outperform their unlabelled counterparts, emphasising the importance of the source domain labelled data for UDA. Moreover, pointwise mutual information and frequency-based pivot selection strategies obtain the best performances in two state-of-the-art UDA methods.

Agent communication languages (ACLs) are fundamental mechanisms that enable agents in multi-agent systems to talk, communicate with each other in order to satisfy their individual and social goals in a cooperative and competitive manner. Social approaches are advocated to overcome the shortcomings of ACL semantics delineated by using mental approaches in the figure of agents’ mental notions. Over the last two decades, social commitments have been the subject of considerable research in some of those social approaches as they provide a powerful representation for modeling and reasoning upon multi-agent interactions in the form of mutual contractual obligations. They particularly provide a declarative, flexible, verifiable, and social semantics for ACL messages while respecting agents’ autonomy, heterogeneity, and openness.

In this manuscript, we go through prominent and predominate proposals in the literature to explore the state of the art on how temporal logics can be devoted to define a formal semantics for ACL messages in terms of social commitments and associated actions. We explain each proposal and point out if and how it meets seven crucial criteria, four of them introduced by Munindar P. Singh to have a well-defined semantics for ACL messages. Far from deciding the best proposal, our aim is to present the advantages (strengths) and limitations of those proposals to designers and developers using a concrete running example and to compare between them, so that they can make the best choice with regard to their needs. We explore and evaluate current specification languages and different verification techniques that have been discussed within those proposals to, respectively, specify and verify commitment-based protocols. We also investigate logical languages of actions advocated to specify, model, and execute commitment-based protocols in other contributed proposals. Finally, we suggest some solutions that can contribute to address the identified limitations.

We survey the use and effect of decomposition-based techniques in qualitative spatial and temporal constraint-based reasoning, and clarify the notions of a tree decomposition, a chordal graph, and a partitioning graph, and their implication with a particular constraint property that has been extensively used in the literature, namely, patchwork. As a consequence, we prove that a recently proposed decomposition-based approach that was presented in the study by Nikolaou and Koubarakis for checking the satisfiability of qualitative spatial constraint networks lacks soundness. Therefore, the approach becomes quite controversial as it does not seem to offer any technical advance at all, while results of an experimental evaluation of it in a following work presented in the study by Sioutis become questionable. Finally, we present a particular tree decomposition that is based on the biconnected components of the constraint graph of a given large network, and show that it allows for cost-free utilization of parallelism for a qualitative constraint language that has patchwork for satisfiable atomic networks.

1 Believable interactive embodied agents

Among the goals of research on autonomous agents one important aim is to build believable interactive embodied agents that are apt to application to friendly interfaces in e-commerce, tourist and service query systems, entertainment (e.g. synthetic actors) and education (pedagogical agents, agents for help and instruction to the hearing impaired).

This paper presents a comprehensive review of evolutionary algorithms that learn an ensemble of predictive models for supervised machine learning (classification and regression). We propose a detailed four-level taxonomy of studies in this area. The first level of the taxonomy categorizes studies based on which stage of the ensemble learning process is addressed by the evolutionary algorithm: the generation of base models, model selection, or the integration of outputs. The next three levels of the taxonomy further categorize studies based on methods used to address each stage. In addition, we categorize studies according to the main types of objectives optimized by the evolutionary algorithm, the type of base learner used and the type of evolutionary algorithm used. We also discuss controversial topics, like the pros and cons of the selection stage of ensemble learning, and the need for using a diversity measure for the ensemble’s members in the fitness function. Finally, as conclusions, we summarize our findings about patterns in the frequency of use of different methods and suggest several new research directions for evolutionary ensemble learning.

The success of a software system strongly depends on the ability of turning a precise domain analysis into a concrete architecture. Even if the domain model relies on sound ontological bases, there is often a wide semantic gap between the conceptual model and the concrete components that should reify it. To fill the semantic gap, relevant domain concepts should be engineered by identifying the corresponding architectural abstractions, which can be realized by concrete software components. Space plays a crucial role in many application domains, but surprisingly, related architectural abstractions have not emerged yet. This paper proposes space-related abstractions derived from the application of classical software engineering principles; in particular, the information hiding principle that leads to an operational definition of space. Basic abstractions are refined to deal with architectural aspects. As the underlying software engineering principles are close to principles that underlie the definition of space ontologies, the conjecture is that the proposed space architectural abstractions might be the basis for a formalization in ontological terms.

This paper focuses on coordination middleware for distributed applications based on active documents and XML technologies. First, the paper introduces the main concepts underlying active documents and XML, and identifies the strict relations between active documents and mobile agents (“document agents”). Then, the paper goes into details about the problem of defining a suitable middleware architecture to support coordination activities in applications including active documents and mobile agents, by specifically focusing on the role played by XML technologies in that context. A simple taxonomy is introduced to characterise coordination middleware architectures depending on the way they exploit XML documents in supporting coordination. The characteristics of several middleware infrastructures are then surveyed and evaluated, also with the help of a simple example scenario in the area of distributed workflow management. This analysis enables us to identify the advantages and the shortcomings of the different approaches, and the basic requirements of a middleware for XML-centric applications.

Coordination infrastructures play a central role in the engineering of multiagent systems. Since the advent of agent technology, research on coordination infrastructures has produced a significant number of infrastructures with varying features. In this paper, we review the the state-of-the-art coordination infrastructures with the purpose of identifying open research challenges that next generation coordination infrastructures should address. Our analysis concludes that next generation coordination infrastructures must address a number of challenges: (i) to become socially aware, by facilitating human interaction within a MAS; (ii) to assist agents in their decision making by providing decision support that helps them reduce the scope of reasoning and facilitates the achievement of their goals; and (iii) to increase openness to support on-line, fully decentralised design and execution. Furthermore, we identify some promising approaches in the literature, together with the research issues worth investigating, to cope with such challenges.

A series of temporal reasoning tasks are identified which motivate the consideration and application of temporal logics in artificial intelligence. There follows a discussion of the broad issues involved in modelling time and constructing a temporal logic. The paper then presents a detailed review of the major approaches to temporal logics: first-order logic approaches, modal temporal logics and reified temporal logics. The review considers the most significant exemplars within the various approaches, including logics due to Russell, Hayes and McCarthy, Prior, McDermott, Allen, Kowalski and Sergot. The logics are compared and contrasted, particularly in their treatments of change and action, the roles they seek to fulfil and the underlying models of time on which they rest. The paper concludes with a brief consideration of the problem of granularity—a problem of considerable significance in temporal reasoning, which has yet to be satisfactorily treated in a temporal logic.

Biometric user authentication in large-scale distributed systems involves passive scanners and networked workstations and databases for user data acquisition, processing, and encryption. Unfortunately, traditional biometric authentication systems are prone to several attacks, such as Replay Attacks, Communication Attacks, and Database Attacks. Embedded biometric sensors overcome security limits of conventional software recognition systems, hiding its common attack points. The availability of mature reconfigurable hardware technology, such as field-programmable gate arrays, allows the developers to design and prototype the whole embedded biometric sensors. In this work, two strong and invasive biometric traits, such as fingerprint and iris, have been considered, analyzed, and combined in unimodal and multimodal biometric sensors. Biometric sensor performance has been evaluated using the well-known FVC2002, CASIA, and BATH databases.

Approximate consensus is an important building block for distributed systems, used overtly or implicitly in applications as diverse as formation control, sensor fusion, and synchronization. Laplacian-based consensus, the current dominant approach, is extremely accurate and resilient, but converges slowly. Comparing Laplacian-based consensus to exact consensus algorithms, relaxing the requirements for accuracy and resilience should enable a spectrum of algorithms that incrementally tradeoff accuracy and/or resilience for speed. This manuscript demonstrates that may be so by beginning to populate this spectrum with a new approach to approximate consensus, Power-Law-Driven Consensus (PLD-consensus), which accelerates consensus by sending values across long distances using a self-organizing overlay network. Both a unidirectional and bidirectional algorithm based on this approach are studied. Although both have the same asymptotic O(diameter) convergence time (vs. O(diameter 2) for Laplacian-based), unidirectional PLD-consensus is faster and more resilient than bidirectional PLD-consensus, but exhibits higher variance in the converged value.

In order to be able to specify and design ubiquitous applications, it is necessary to rely on a precise and formal representation of context. Aiming at reducing the gap between the formal specification and the design of a concrete ubiquitous application, we have proposed an approach to specify a context-aware system at three levels of abstraction. This specification uses the Algebra of Contextualized Ontologies, which is based on Category Theory and takes contextualization as a basic notion, proposing a small set of simple and powerful operations to compose and decompose contextualized ontologies. In this paper, we present a revised version of the algebra and algorithms to compute the algebraic operations. We exemplify our approach along with a concrete ubiquitous computing scenario. Moreover, by taking advantage of Category Theory foundations, we show that it is possible to derive the algorithms for each algebraic operation from simple modifications of a single basic algorithm.

Both the advent of large communication networks and the development of agent models designing more and more sophisticated (in some cases autonomous) behaviours, are producing what can be called an Agent Society. Agent societies are constituted by both artificial and human agents (in perspective indistinguishable from each other) communicating with each other (not necessarily each with all the others), with individual or collective tasks, different resources, different skills, and so on. In other words, these agent societies will become more and more similar to the human ones. As in real societies, in these virtual societies it will be necessary to consider all the problems connected with secure and reliable interaction, with trusting the other agents, with the possibility to be frauded, cheated, deceived in some exchange or relationship.

Analogues to much of today's work in ontologies have existed for centuries in text retrieval. The use of controlled vocabularies, or thesauri, has been fundamental to document indexing in library science. Thesauri serve several purposes, including:

[bull ] Knowledge organisation A thesaurus provides a hierarchy of concepts that organises domain-specific knowledge.

[bull ] Terminology normalisation By selecting a unique word or phrase to represent each domain concept, then linking synonymous terms to it, a thesaurus enforces terminological consistency.

[bull ] Query expansion A thesaurus facilitates the addition of terms to a query by providing explicit hierarchical and lateral relationships among terms.

These properties serve to mediate the information flow from indexer to user. Thesauri thus serve many of the same functions for people that ontologies are designed to serve for software agents. As automated retrieval has developed over the decades since the inception of computer processing of text, many techniques have been introduced to apply this typically manual work to the automated arena (see Soergel (1985) for an introduction to library information systems, also Anderson and Pélrez-Carballo (2001a, 2001b) for a summary of the intersection of human and machine indexing).

Designing is a skill central to many human tasks. Designers are constantly producing newer and better artifacts, generating innovative solutions to problems in our world. This paper looks at innovation, and research that is aimed at developing theories and methodologies for innovative design. We view design as a process of association and exploration. These two approaches are fundamental to innovation. The aim of exploration is to generate a large variety of design alternatives by breaking away from the norms by looking in unlikely places, and by relaxing binding constraints. Exploration exposes possibilities that would not normally hâve been considered, possibilities that may serendipitously lead to innovative solutions. Association, on the other hand, attempts to exploit previous design experiences in a new design context. This is donc by recognizing useful analogies that can help in synthesizing parts of a design, recognizing unforeseen problems, and discovering opportunities. This paper is the first part of a two-part paper that presents and discusses a variety of association and exploration methods. This part examines association-based techniques, some of which have been used in actual design Systems, and others that point to the solution of some open questions in design research. We develop these ideas by examining connections between design research and other disciplines such as cognitive psychology, artificial intelligence, the history of science, and the creativity literature.

This paper addresses the problem of implementing a Simultaneous Localization and Mapping (SLAM) algorithm combined with a non-reactive controller (such as trajectory following or path following). A general study showing the advantages of using predictors to avoid mapping inconsistences in autonomous SLAM architectures is presented. In addition, this paper presents a priority-based uncertainty map construction method of the environment by a mobile robot when executing a SLAM algorithm. The SLAM algorithm is implemented with an extended Kalman filter (EKF) and extracts corners (convex and concave) and lines (associated with walls) from the surrounding environment. A navigation approach directs the robot motion to the regions of the environment with the higher uncertainty and the higher priority. The uncertainty of a region is specified by a probability characterization computed at the corresponding representative points. These points are obtained by a Monte Carlo experiment and their probability is estimated by the sum of Gaussians method, avoiding the time-consuming map-gridding procedure. The priority is determined by the frame in which the uncertainty region was detected (either local or global to the vehicle's pose). The mobile robot has a non-reactive trajectory following controller implemented on it to drive the vehicle to the uncertainty points. SLAM real-time experiments in real environment, navigation examples, uncertainty maps constructions along with algorithm strategies and architectures are also included in this work.

Today, management and tuning questions are approached using if… then… rules of thumb. This reactive approach requires expertise regarding system behavior, making it difficult to deal with unforeseen uses of a system’s resources and leading to system unpredictability and large system management overheads. We propose a What…if… approach that allows interactive exploration of the effects of system changes, thus converting complex tuning problem into simpler search problems. Through two concrete management problems, automating system upgrades and deciding on service migrations, we identify system design changes that enable a system to answer What…if… questions about itself.

Loyalty cards programs have been used by retailers to increase customer retention. Loyality cards provide means to identify a particular customer and to collect customer-specific data, thus enabling individualized marketing; however, operating a loyalty program is complicated for retailers since they require to manage balances, collections, and transfers of customers. This is exactly the same problem the retailers were facing before credit cards were readily available. A new problem is that customers now have too many cards, customers may forget, or even deliberately decide to carry only a selection of their cards. This paper proposes a loyalty program based on a blockchain that does not require a physical card for identifying customers as it associates customers to their phone numbers, since nowadays people always carry their phone. In this perspective, companies can reduce overhead costs associated to managing the loyalty program. This paper reviews the technology required and describes the implementation of a loyalty program based on blockchains. Finally, it also enumerates the reasons for choosing the blockchain technology for this application.

This paper presents a parameterized gait generator based on linear inverted pendulum model (LIPM) theory, which allows users to generate a natural gait pattern with desired step sizes. Five types of zero moment point (ZMP) components are proposed for formulating a natural ZMP reference, where ZMP moves continuously during single support phases instead of staying at a fixed point in the sagittal and lateral plane. The corresponding center of mass (CoM) trajectories for these components are derived by LIPM theory. To generate a parameterized gait pattern with user-defined parameters, a gait planning algorithm is proposed, which determines related coefficients and boundary conditions of the CoM trajectory for each step. The proposed parameterized gait generator also provides a concept for users to generate gait patterns with self-defined ZMP references by using different components. Finally, the feasibility of the proposed method is validated by the experimental results with a teen-sized humanoid robot, David, which won first place in the sprint event at the 20th Federation of International Robot-soccer Association (FIRA) RoboWorld Cup.