Through the use of rapid prototyping it is possible to move directly from a computer design to a nal physical product without human intervention. Telemanufacturing is an extension to rapid prototyping that presents a data communications based interface to rapid prototyping. Resource negotiating agents are able to provide a completely autonomous negotiation process. By integrating resource negotiating agents into a telemanufacturing environment, an opportunity for a completely automated supply chain is presented. / Prof. E.M. Ehlers

Current syndromic surveillance systems utilize centralized databases that are neither scalable in storage space nor in computing power. Such systems are limited in the amount of syndromic data that may be collected and analyzed for the early detection of infectious disease outbreaks. However, with the increased prevalence of international travel, public health monitoring must extend beyond the borders of municipalities or states which will require the ability to store vasts amount of data and significant computing power for analyzing the data. Intelligent mobile agents may be used to create a distributed surveillance system that will utilize the hard drives and computer processing unit (CPU) power of the hosts on the agent network where the syndromic information is located. This thesis proposes the design of a mobile agent-based syndromic surveillance system and an agent decision model for outbreak detection. Simulation results indicate that mobile agents are capable of detecting an outbreak that occurs at all hosts the agent is monitoring. Further study of agent decision models is required to account for localized epidemics and variable agent movement rates.

No description available.

Thesis (MScEng)--Stellenbosch University, 2013. / ENGLISH ABSTRACT: The thesis evaluates control strategies for reconfigurable manufacturing systems by using a welding assembly cell as a case study. The cell consists of a pallet magazine, conveyor, feeder subsystem (comprising an articulated robot and singulation unit), welder subsystem (which uses a modular Cartesian robot), and inspection and removal subsystems. The research focuses on control strategies that enhance reconfigurability in terms of structure, hardware and software using agent-based control and the IEC 61499 standard, based on PC control. Reconfiguration may occur when a new product is introduced, as well as when a new subsystem is introduced or removed from the production cell. The overall control architecture is that the subsystems retain no knowledge of the product, but product information resides in the cell controller, while services offered by the subsystems are registered with the directory facilitator of the Java agent platform. The control strategies are implemented on the modular Cartesian weld robot and the cell controller for assembly cell. A layered architecture with low-level control and high-level control is used to allow separation of concerns and rapid changes in both hardware and software components. The low-level control responds in hard real-time to internal and external events, while the high-level control handles soft real-time actions involving coordination of control related issues. The results showed IEC 61499 function blocks to be better suited to low-level control application in distributed systems, while agents are more suited for high-level control. Modularity in software components enhances hardware and software scalability. Additionally, agents can support online reconfiguration of reconfigurable machines. / AFRIKAANSE OPSOMMING: Die tesis evalueer beheerstrategieë vir herkonfigureerbare vervaardigingstelsels deur gebruik te maak van ’n sweismonteersel as ’n gevallestudie. Die sel bestaan uit ’n palletmagasyn, vervoerbande, voersubstelsel (bestaande uit ’n geartikuleerde robot en singulasie-eenheid), sweissubstelsel (wat gebruik maak van ’n modulêre Cartesiese robot), en inspeksie- en verwyderingsubstelsels. Die navorsing fokus op beheerstrategieë wat herkonfigureerbaarheid verhoog in terme van struktuur, hardeware en sagteware met behulp van agent-gebaseerde beheer en die IEC 61499 standaard, wat gebaseer is op PC-beheer. Herkonfigurasie mag voorkom wanneer ’n nuwe produk in-gestel word, sowel as wanneeer ’n nuwe substelsel bygevoeg of verwyder word van die produksiesel. Die oorhoofse beheerargitektuur is dat die substelsels geen kennis van die produk hou nie, maar die produkinligting in die selbeheerder geberg, terwyl dienste wat aangebied word deur die substelsels wat geregistreer is by die gidsfasiliteerder van die Java agent platform. Die beheerstrategië is geïmplementeer op die modulere Cartesiese sweisrobot en die selbeheerder vir die monteersel. ’n Gelaagde argitektuur met ’n lae-vlak beheer en hoë-vlak beheer word gebruik om skeiding van oorwegings en vinnige veranderinge in beide hardeware en sagteware komponente toe te laat. Die lae-vlak beheer reageer hard intyds op interne en eksterne gebeure, terwyl die hoë-vlak beheer sag intyds die koördinering van beheerverwante kwessies hanteer. Die resultate het getoon dat IEC 61499 funksie-blokke beter geskik is vir lae-vlak beheer toepassing in verspreide stelsels, terwyl agente meer geskik is vir hoë-vlak beheer. Modulariteit in sagteware komponente verhoog hardeware en sagteware skaleerbaarheid. Boonop kan agente ook aanlyn herkonfigurasie van herkonfigureerbare masjiene ondersteun.

One of the hardest challenges in the field of machine learning is to build agents, such as robotic assistants in homes and hospitals, that can autonomously learn new tasks that they were not pre-programmed to tackle, without the intervention of an engineer. Reinforcement learning (RL) and learning from demonstration (LfD) are popular approaches for task learning, but they are often ineffective in high-dimensional domains unless provided with either a great deal of problem-specific domain information or a carefully crafted representation of the state and dynamics of the world. Unfortunately, autonomous agents trying to learn new tasks usually do not have access to such domain information nor to an appropriate representation. We demonstrate that algorithms that focus, at each moment, on the relevant features of the state space can achieve significant speed-ups over previous reinforcement learning algorithms with respect to the number of state features in complex domains. To do so, we introduce and evaluate a family of attention focus algorithms. We show that these algorithms can reduce the dimensionality of complex domains, creating a compact representation of the state space with which satisficing policies can be learned efficiently. Our approach obtains exponential speed-ups with respect to the number of features considered when compared with table-based learning algorithms and polynomial speed-ups when compared with state-of-the-art function approximation RL algorithms such as LSPI or fitted Q-learning. Our attention focus algorithms are divided in two classes, depending on the source of the focus information they require. Attention focus from human demonstrations infers the features to focus on from a set of demonstrations from human teachers performing the task the agent must learn. We introduce two algorithms within this class. The first one, abstraction from demonstration (AfD), identifies features that can be safely ignored in the whole state space and builds a state-space abstraction where a satisficing policy can be learned efficiently. The second, automatic decomposition and abstraction from demonstration, goes one step further, using the demonstrations to identify a set of subtasks and to find an appropriate abstraction for each subtask found. The other class of algorithms we present, attention focus with a world model, does not require a set of human demonstrations. Instead, it extracts the attention focus information from an object-based model of the world together with the agent experience in performing the task. Within this class, we introduce object-focused Q-learning (OF-Q), at first with an assumption of object independence that is later removed to support domains where objects interact with each other. Finally, we show that both sources of focus information can be combined for further speed-ups.

Changes in work, along with improvements in techniques to statistically model uncertainty, have resulted in a class of groupware tools able to forecast the activities and/or attentional state of their users. This thesis represents an exploration into the design, development, and use of one such system. I describe the design and development of a groupware calendar system called Augur that is augmented with the ability to predict the attendance of its users. Using Bayesian networks, Augur models the uncertain problem of event attendance, drawing inferences based on the attributes of calendar events as well as a history of attendance provided by each user. This system was deployed to an academic workgroup and studied over the course of a semester. To more deeply explore the social implications of Augur and systems like it, I conducted a structured privacy analysis of Augur to examine the vulnerabilities inherent in this type of forecasting groupware system. I present an architecture, user interface, and probabilistic model for Augur. This work also addresses the feasibility of such a system and the challenges faced when deploying it to an academic workgroup. I also report on an exploration of the systems use by individuals, its effects on communication within working relationships, and its effectiveness with respect to the presence of domestic calendars. Finally, I present a set of implications for the workplace social environment with the introduction of Augur. Specifically, I show how the integrity of predictions generated by Augur can have consequences for the privacy of users and their representations through the shared calendar. Overall, this thesis is presented as an early exploration into the potential for a new class of forecasting groupware applications. It offers guidance and lessons learned for both designers and researchers seeking to work in this area. It also presents a complete calendar application as an example for building and studying such systems.

In this work, I view incremental experiential learning in intelligent software agents as progressive agent self-adaptation. When an agent produces an incorrect behavior, then it may reflect on, and thus diagnose and repair, the reasoning and knowledge that produced the incorrect behavior. In particular, I focus on the self-diagnosis and self-repair of an agent's domain knowledge. The implementation of systems with the capability to self-diagnose and self-repair involves building both reasoning processes capable of such learning and knowledge representations capable of supporting those reasoning processes. The core issue my dissertation addresses is: what kind of metaknowledge (knowledge about knowledge) may enable the agent to diagnose faults in its domain knowledge? In providing a solution to this issue, the central contribution of this research is a theory of the kind of metaknowledge that enables a system to reason about and adapt its conceptual knowledge. For this purpose, I propose a representation that explicitly encodes metaknowledge in the form of procedures called Empirical Verification Procedures (EVPs). In the proposed knowledge representation, an EVP is associated with each concept within the agent's domain knowledge. Each EVP explicitly semantically grounds the associated concept in the agent's perception, and can thus be used as a test to determine the validity of knowledge of that concept during diagnosis. I present the formal and empirical evaluation of a system, Augur, that makes use of EVP metaknowledge to adapt its own domain knowledge in the context of a particular subclass of classification problem that I call compositional classification, in which the overall classification task can be broken into a hierarchically organized set of subtasks. I hypothesize that EVP metaknowledge will enable a system to automatically adapt its knowledge in two ways: first, by adjusting the ways that inputs are categorized by a concept, in accordance with semantics fixed by an associated EVP; and second, by adjusting the semantics of concepts themselves when they fail to contribute appropriately to system goals. The latter adaptation is realized by altering the EVP associated with the concept in question. I further hypothesize that the semantic grounding of domain concepts in perception through the use of EVPs will increase the generalization power of a learner that operates over those concepts, and thus make learning more efficient. Beyond the support of these hypotheses, I also present results pertinent to the understanding of learning in compositional classification settings using structured knowledge representations.

Texte remanié de : Dissertation : Droit : Université de Hambourg : 2008. / Bibliogr. p. 225-252. Notes bibliogr.

Thesis (Ph.D.)--University of Melbourne, Dept. of Computer Science and Software Engineering, 2006. / Typescript. Includes bibliographical references (leaves 291-312).

Thesis (Ph. D.)--University of Karlsruhe, 2001. / Includes bibliographical references (p. [193]-201) and subject index. Also available in electronic format on the Internet.