Sistema multiagente web semântico para gestão de conteúdos educacionais

Pós-graduação em Ciência da Computação - IBILCE

Uma arquitetura multiagente para sistemas de supervisão e controle de processos

Pós-graduação em Ciência da Computação - IBILCE

Arquitetura de um sistema multiagente autônomo para supervisão e controle

This paper presents a multi-agent architecture that was designed to develop processes supervision and control systems, with the main objective to automate tasks that are repetitive and stressful, and error prone when performed by humans. A set of agents were identified, based on the study of a number of applications found in the literature, that use the approach of multi-agent systems for data integration and process monitoring to faults detection and diagnosis, these agents are used as basis of the proposed multi-agent architecture. A prototype system for the analysis of abnormalities during oil wells drilling was developed.

A communication infrastructure to support knowledge level agents on the web

Agent Communication Languages (ACLs) have been developed to provide a way for agents to communicate with each other supporting cooperation in Multi-Agent Systems. In the past few years many ACLs have been proposed for Multi-Agent Systems, such as KQML and FIPA-ACL. The goal of these languages is to support high-level, human like communication among agents, exploiting Knowledge Level features rather than symbol level ones. Adopting these ACLs, and mainly the FIPA-ACL specifications, many agent platforms and prototypes have been developed. Despite these efforts, an important issue in the research on ACLs is still open and concerns how these languages should deal (at the Knowledge Level) with possible failures of agents. Indeed, the notion of Knowledge Level cannot be straightforwardly extended to a distributed framework such as MASs, because problems concerning communication and concurrency may arise when several Knowledge Level agents interact (for example deadlock or starvation). The main contribution of this Thesis is the design and the implementation of NOWHERE, a platform to support Knowledge Level Agents on the Web. NOWHERE exploits an advanced Agent Communication Language, FT-ACL, which provides high-level fault-tolerant communication primitives and satisfies a set of well defined Knowledge Level programming requirements. NOWHERE is well integrated with current technologies, for example providing full integration for Web services. Supporting different middleware used to send messages, it can be adapted to various scenarios. In this Thesis we present the design and the implementation of the architecture, together with a discussion of the most interesting details and a comparison with other emerging agent platforms. We also present several case studies where we discuss the benefits of programming agents using the NOWHERE architecture, comparing the results with other solutions. Finally, the complete source code of the basic examples can be found in appendix.

Specification and Verification of Declarative Open Interaction Models - A Logic-based framework

The advent of distributed and heterogeneous systems has laid the foundation for the birth of new architectural paradigms, in which many separated and autonomous entities collaborate and interact to the aim of achieving complex strategic goals, impossible to be accomplished on their own. A non exhaustive list of systems targeted by such paradigms includes Business Process Management, Clinical Guidelines and Careflow Protocols, Service-Oriented and Multi-Agent Systems. It is largely recognized that engineering these systems requires novel modeling techniques. In particular, many authors are claiming that an open, declarative perspective is needed to complement the closed, procedural nature of the state of the art specification languages. For example, the ConDec language has been recently proposed to target the declarative and open specification of Business Processes, overcoming the over-specification and over-constraining issues of classical procedural approaches. On the one hand, the success of such novel modeling languages strongly depends on their usability by non-IT savvy: they must provide an appealing, intuitive graphical front-end. On the other hand, they must be prone to verification, in order to guarantee the trustworthiness and reliability of the developed model, as well as to ensure that the actual executions of the system effectively comply with it. In this dissertation, we claim that Computational Logic is a suitable framework for dealing with the specification, verification, execution, monitoring and analysis of these systems. We propose to adopt an extended version of the ConDec language for specifying interaction models with a declarative, open flavor. We show how all the (extended) ConDec constructs can be automatically translated to the CLIMB Computational Logic-based language, and illustrate how its corresponding reasoning techniques can be successfully exploited to provide support and verification capabilities along the whole life cycle of the targeted systems.

Mobile TuCSoN: theoretical and technological requirements for TuCSoN's porting over Android mobile devices

Communication and coordination are two key-aspects in open distributed agent system, being both responsible for the system’s behaviour integrity. An infrastructure capable to handling these issues, like TuCSoN, should to be able to exploit modern technologies and tools provided by fast software engineering contexts. Thesis aims to demonstrate TuCSoN infrastructure’s abilities to cope new possibilities, hardware and software, offered by mobile technology. The scenarios are going to configure, are related to the distributed nature of multi-agent systems where an agent should be located and runned just on a mobile device. We deal new mobile technology frontiers concerned with smartphones using Android operating system by Google. Analysis and deployment of a distributed agent-based system so described go first to impact with quality and quantity considerations about available resources. Engineering issue at the base of our research is to use TuCSoN against to reduced memory and computing capability of a smartphone, without the loss of functionality, efficiency and integrity for the infrastructure. Thesis work is organized on two fronts simultaneously: the former is the rationalization process of the available hardware and software resources, the latter, totally orthogonal, is the adaptation and optimization process about TuCSoN architecture for an ad-hoc client side release.

BDI agents for Real Time Strategy games

While the use of distributed intelligence has been incrementally spreading in the design of a great number of intelligent systems, the field of Artificial Intelligence in Real Time Strategy games has remained mostly a centralized environment. Despite turn-based games have attained AIs of world-class level, the fast paced nature of RTS games has proven to be a significant obstacle to the quality of its AIs. Chapter 1 introduces RTS games describing their characteristics, mechanics and elements. Chapter 2 introduces Multi-Agent Systems and the use of the Beliefs-Desires-Intentions abstraction, analysing the possibilities given by self-computing properties. In Chapter 3 the current state of AI development in RTS games is analyzed highlighting the struggles of the gaming industry to produce valuable. The focus on improving multiplayer experience has impacted gravely on the quality of the AIs thus leaving them with serious flaws that impair their ability to challenge and entertain players. Chapter 4 explores different aspects of AI development for RTS, evaluating the potential strengths and weaknesses of an agent-based approach and analysing which aspects can benefit the most against centralized AIs. Chapter 5 describes a generic agent-based framework for RTS games where every game entity becomes an agent, each of which having its own knowledge and set of goals. Different aspects of the game, like economy, exploration and warfare are also analysed, and some agent-based solutions are outlined. The possible exploitation of self-computing properties to efficiently organize the agents activity is then inspected. Chapter 6 presents the design and implementation of an AI for an existing Open Source game in beta development stage: 0 a.d., an historical RTS game on ancient warfare which features a modern graphical engine and evolved mechanics. The entities in the conceptual framework are implemented in a new agent-based platform seamlessly nested inside the existing game engine, called ABot, widely described in Chapters 7, 8 and 9. Chapter 10 and 11 include the design and realization of a new agent based language useful for defining behavioural modules for the agents in ABot, paving the way for a wider spectrum of contributors. Chapter 12 concludes the work analysing the outcome of tests meant to evaluate strategies, realism and pure performance, finally drawing conclusions and future works in Chapter 13.

Enabling a direct path from end-user specifications to executable protocols in a biology laboratory environment

Biomedical analyses are becoming increasingly complex, with respect to both the type of the data to be produced and the procedures to be executed. This trend is expected to continue in the future. The development of information and protocol management systems that can sustain this challenge is therefore becoming an essential enabling factor for all actors in the field. The use of custom-built solutions that require the biology domain expert to acquire or procure software engineering expertise in the development of the laboratory infrastructure is not fully satisfactory because it incurs undesirable mutual knowledge dependencies between the two camps. We propose instead an infrastructure concept that enables the domain experts to express laboratory protocols using proper domain knowledge, free from the incidence and mediation of the software implementation artefacts. In the system that we propose this is made possible by basing the modelling language on an authoritative domain specific ontology and then using modern model-driven architecture technology to transform the user models in software artefacts ready for execution in a multi-agent based execution platform specialized for biomedical laboratories.

Validation of the CFD code NEPTUNE for a full scale simulator for decay heat removal systems with in-pool heat exchangers

In the present work, a multi physics simulation of an innovative safety system for light water nuclear reactor is performed, with the aim to increase the reliability of its main decay heat removal system. The system studied, denoted by the acronym PERSEO (in Pool Energy Removal System for Emergency Operation) is able to remove the decay power from the primary side of the light water nuclear reactor through a heat suppression pool. The experimental facility, located at SIET laboratories (PIACENZA), is an evolution of the Thermal Valve concept where the triggering valve is installed liquid side, on a line connecting two pools at the bottom. During the normal operation, the valve is closed, while in emergency conditions it opens, the heat exchanger is flooded with consequent heat transfer from the primary side to the pool side. In order to verify the correct system behavior during long term accidental transient, two main experimental PERSEO tests are analyzed. For this purpose, a coupling between the mono dimensional system code CATHARE, which reproduces the system scale behavior, with a three-dimensional CFD code NEPTUNE CFD, allowing a full investigation of the pools and the injector, is implemented. The coupling between the two codes is realized through the boundary conditions. In a first analysis, the facility is simulated by the system code CATHARE V2.5 to validate the results with the experimental data. The comparison of the numerical results obtained shows a different void distribution during the boiling conditions inside the heat suppression pool for the two cases of single nodalization and three volume nodalization scheme of the pool. Finaly, to improve the investigation capability of the void distribution inside the pool and the temperature stratification phenomena below the injector, a two and three dimensional CFD models with a simplified geometry of the system are adopted.

Coordination and Control of Autonomous Mobile Robots Swarms by using Particle Swarm Optimization Algorithm and Consensus Theory

This thesis presents some different techniques designed to drive a swarm of robots in an a-priori unknown environment in order to move the group from a starting area to a final one avoiding obstacles. The presented techniques are based on two different theories used alone or in combination: Swarm Intelligence (SI) and Graph Theory. Both theories are based on the study of interactions between different entities (also called agents or units) in Multi- Agent Systems (MAS). The first one belongs to the Artificial Intelligence context and the second one to the Distributed Systems context. These theories, each one from its own point of view, exploit the emergent behaviour that comes from the interactive work of the entities, in order to achieve a common goal. The features of flexibility and adaptability of the swarm have been exploited with the aim to overcome and to minimize difficulties and problems that can affect one or more units of the group, having minimal impact to the whole group and to the common main target. Another aim of this work is to show the importance of the information shared between the units of the group, such as the communication topology, because it helps to maintain the environmental information, detected by each single agent, updated among the swarm. Swarm Intelligence has been applied to the presented technique, through the Particle Swarm Optimization algorithm (PSO), taking advantage of its features as a navigation system. The Graph Theory has been applied by exploiting Consensus and the application of the agreement protocol with the aim to maintain the units in a desired and controlled formation. This approach has been followed in order to conserve the power of PSO and to control part of its random behaviour with a distributed control algorithm like Consensus.

Teoria del consenso e applicazione al problema del coordinamento del moto di robot.

Il sempre crescente numero di applicazioni di reti di sensori, robot cooperanti e formazioni di veicoli, ha fatto sì che le problematiche legate al coordinamento di sistemi multi-agente (MAS) diventassero tra le più studiate nell’ambito della teoria dei controlli. Esistono numerosi approcci per affrontare il problema, spesso profondamente diversi tra loro. La strategia studiata in questa tesi è basata sulla Teoria del Consenso, che ha una natura distribuita e completamente leader-less; inoltre il contenuto informativo scambiato tra gli agenti è ridotto al minimo. I primi 3 capitoli introducono ed analizzano le leggi di interazione (Protocolli di Consenso) che permettono di coordinare un Network di sistemi dinamici. Nel capitolo 4 si pensa all'applicazione della teoria al problema del "loitering" circolare di più robot volanti attorno ad un obiettivo in movimento. Si sviluppa a tale scopo una simulazione in ambiente Matlab/Simulink, che genera le traiettorie di riferimento di raggio e centro impostabili, a partire da qualunque posizione iniziale degli agenti. Tale simulazione è stata utilizzata presso il “Center for Research on Complex Automated Systems” (CASY-DEI Università di Bologna) per implementare il loitering di una rete di quadrirotori "CrazyFlie". I risultati ed il setup di laboratorio sono riportati nel capitolo 5. Sviluppi futuri si concentreranno su algoritmi locali che permettano agli agenti di evitare collisioni durante i transitori: il controllo di collision-avoidance dovrà essere completamente indipendente da quello di consenso, per non snaturare il protocollo di Consenso stesso.

Spike-based decision learning of nash equilibria in two-player games

Humans and animals face decision tasks in an uncertain multi-agent environment where an agent's strategy may change in time due to the co-adaptation of others strategies. The neuronal substrate and the computational algorithms underlying such adaptive decision making, however, is largely unknown. We propose a population coding model of spiking neurons with a policy gradient procedure that successfully acquires optimal strategies for classical game-theoretical tasks. The suggested population reinforcement learning reproduces data from human behavioral experiments for the blackjack and the inspector game. It performs optimally according to a pure (deterministic) and mixed (stochastic) Nash equilibrium, respectively. In contrast, temporal-difference(TD)-learning, covariance-learning, and basic reinforcement learning fail to perform optimally for the stochastic strategy. Spike-based population reinforcement learning, shown to follow the stochastic reward gradient, is therefore a viable candidate to explain automated decision learning of a Nash equilibrium in two-player games.

Control concept for autonomous changeable material flow systems

Today’s material flow systems for mass customization or dynamic productions are usually realized with manual transportation systems. However new concepts in the domain of material flow and device control like function-oriented modularization and intelligent multi-agent-systems offer the possibility to employ changeable and automated material flow systems in dynamic production structures. These systems need the ability to react on unplanned and unexpected events autonomously.

Agentenbasiertes Planungsmodell für die Grobplanung von Kommissioniersystemen

Rechnergestützte Modellansätze, die Logistiksysteme gestalten und generieren, sind eine hochkomplexe Aufgabenstellung. Die bisher in der Praxis existierenden Planungs- und Steuerungsmodelle für Intralogistiksysteme weisen für die aktuellen und zukünftigen Anforderungen wie der Komplexitätsbewältigung, Reaktionsschnelligkeit und Anpassungsfähigkeit Schwachstellen auf. – Ein innovativer Ansatz, diesen Ansprüchen gerecht zu werden, stellen Multiagentensysteme dar. Mit ihrem dezentralen und modularen Charakter sind sie für ein komplexes Problem mit einem geringen Grad an Strukturiertheit geeignet. Außerdem ermöglichen diese computergestützten intelligenten Systeme den Anwendern eine einfache und aufwandsarme Handhabung.

Dynamische Standsicherheit von Portalstaplern

Der Artikel beschreibt die Untersuchung der dynamischen Standsicherheit von Portalstaplern. Dazu wurde ein aktueller repräsentativer Portalstapler analysiert und in ein Mehrkörper-Simulationsmodell abgebildet. Die Validierung des Modells erfolgte auf Basis von Messfahrten am realen System. Aus den durchgeführten Simulationsstudien wurden technische Verbesserungsmaßnahmen abgeleitet, welche die dynamische Standsicherheit von Portalstaplern erhöhen.