Manufacturers are facing increasing pressure to reduce the development costs and deployment times for automated assembly systems. This is especially true for a variety of precision mechatronic products. To meet new and changing market needs, the difficulties of integrating their systems must be significantly reduced. Since 1994, the Microdynamic Systems Laboratory at Carnegie Mellon University has been developing an automation framework, called Agile Assembly Architecture (AAA). Additionally to the concept, a prototype instantiation, in the form of a modular tabletop precision assembly system termed Minifactory, has been developed. The platform, provided by the Minifactory and AAA, is able to support and integrate various precision manufacturing processes. These are needed to assemble a large variety of small mechatronic products. In this thesis various enhancements for a second generation agent-based micro assembly system are designed, implemented, tested and improved. The project includes devising methods for tray feeding of precision high-value parts, micro fastening techniques and additional work on visual- and force-servoing. To help achieving these functions, modular and reconfigurable robot end-effectors for handling millimeter sized parts have been designed and built for the existing robotic agents. New concepts for robot end effectors to grasp and release tiny parts, including image processing and intelligent control software, were required and needed to be implemented in the prototype setup. These concepts need to distinguish themselves largely from traditional handling paradigms, in order to solve problems introduced by electrostatic and surface tension forces, that are dominant in manipulating parts that are millimeter and less in size. In order to have a modular system, the factory the main part of this project was the initialization and auto calibration of the different agents. The main focus, of this research, is on improving the design, deployment and reconfiguration capabilities of automated assembly systems for precision mechatronic products. This helps to shorten the development process as well as the assembly of factory systems.  A strategic application for this approach is the automated assembly of small sensors, actuators, medical devices and chip-scale atomic systems such as atomic clocks, magnetometers and gyroscopes.

Networks of communicating agents require distributed algorithms for a variety of tasks in the field of network analysis and control. For applications such as swarms of autonomous vehicles, ad hoc and wireless sensor networks, and such military and civilian applications as exploring and patrolling a robust autonomous system that uses a distributed algorithm for self-partitioning can be significantly helpful. A single team of autonomous vehicles in a field may need to self-dissemble into multiple teams, conducive to completing multiple control tasks. Moreover, because communicating agents are subject to changes, namely, addition or failure of an agent or link, a distributed or decentralized algorithm is favorable over having a central agent. A framework to help with the study of self-partitioning of such multi agent systems that have most basic mobility model not only saves our time in conception but also gives us a cost effective prototype without negotiating the physical realization of the proposed idea. In this thesis I present my work on the implementation of a flexible and distributed stochastic partitioning algorithm on the Lego® Mindstorms’ NXT on a graphical programming platform using National Instruments’ LabVIEW™ forming a team of communicating agents via NXT-Bee radio module. We single out mobility, communication and self-partition as the core elements of the work. The goal is to randomly explore a precinct for reference sites. Agents who have discovered the reference sites announce their target acquisition to form a network formed based upon the distance of each agent with the other wherein the self-partitioning begins to find an optimal partition. Further, to illustrate the work, an experimental test-bench of five Lego NXT robots is presented.

We introduce a generic simulation framework suitable for agent-based simulations featuring the support of heterogeneous agents, hierarchical scheduling and flexible specification of design parameters. One key aspect of this framework is the design specification: we use an XML-based format which is simple-structured yet still enables the design of flexible models. Another issue in agent-based simulations, especially when ready-made components are used, is the heterogeneity arising from both the agents' implementations and the underlying platforms. To tackle these kind of obstacles, we introduce a wrapper technique for mapping the functionality of agents living in an interpreter-based environment to a standardized JAVA interface, thus facilitating the task for any control mechanism (like a simulation manager) because it has to handle only one set of commands for all agents involved. Again, this mapping is made by an XML-based definition format. We demonstrate the technique by applying it to a simple sample simulation of two mass marketing firms operating in an artificial consumer environment. / Series: Report Series SFB "Adaptive Information Systems and Modelling in Economics and Management Science"

Mémoire numérisé par la Direction des bibliothèques de l'Université de Montréal.

Business agency agreement Abstract The purpose of my thesis is to analyse the business agency agreement and describe its essential characteristics. The reason for my research is the practical aplication of this agreement, which enables a developement of business and businessman is provided with an oportunity of expansion. This thesis is divided into twelve chapters and each chapter is dealing with diferent aspects of business agency agreement. Chapter One is introductory and describes main goals of this thesis and essential characteristics of business agency agreement. Chapter Two is subdivided into three parts and focuses on historical development of representation. First part describes representation influenced by Roman law, part two focuses on representation and agency during 19th and 20th century and part two investigates diference between direct and indirect representation, which is necessary for understanding of the business agency agreement. Chapter Three describes the term business agency agreement and its essential elements. It is subdivided into two parts. First one focuces on diference between business agency agreement and its differentiation from employment contract. Second one describes elements such as contarcting parties (the agent and the pricipal) and obligation between them, characterizes...

This project first explored various methods of designing a communication and control system for an AGV. It then implemented a SOA based communication system in ROS on the selected AGV. The ROS package created in the project implemented functions of the Aria and ArNetworking libraries from Adept. The next step of the project implemented the functions of teleoperation, mapping and transfer of maps and navigation from Aria into ROS. The packages built implemented these functions in different ways to test the best method to transfer Aria functions into ROS. A generic set of rules were then formulated that aided the conversion of these functions for users unfamiliar with either of the two packages (ROS and Aria).

No description available.

In der vorliegenden Arbeit wird das Problem der Kalibrierung Agenten-basierter Simulationen (ABS) behandelt, also das Problem, die Parameterwerte eines Agenten-basierten Simulationsmodells so einzustellen, dass valides Simulationsverhalten erreicht wird. Das Kalibrierungsproblem für Simulationen an sich ist nicht neu und ist im Rahmen klassischer Simulationsparadigmen, wie z.B. der Makro-Simulation, fester Bestandteil der Forschung. Im Vergleich zu den dort betrachteten Kalibrierungsproblemen zeichnet sich das Kalibrierungsproblem für ABS jedoch durch eine Reihe zusätzlicher Herausforderungen aus, welche die direkte Anwendung existierender Kalibrierungsverfahren in begrenzter Zeit erschweren, bzw. nicht mehr sinnvoll zulassen. Die Lösung dieser Probleme steht im Zentrum dieser Dissertation: Das Ziel besteht darin, den Nutzer bei der Kalibrierung von ABS auf der Basis von unzureichenden, potentiell fehlerhaften Daten und Wissen zu unterstützen. Dabei sollen drei Hauptprobleme gelöst werden: 1)Vereinfachung der Kalibrierung großer Agenten-Parametermengen auf der Mikro- Ebene in Agenten-basierten Simulationen durch Ausnutzung der spezifischen Struktur von ABS (nämlich dem Aufbau aus einer Menge von Agentenmodellen). 2)Kalibrierung Agenten-basierter Simulationen, so dass auf allen relevanten Beobachtungsebenen valides Simulationsverhalten erzeugt wird (mindestens Mikro und Makro-Ebene). Als erschwerende Randbedingung muss die Kalibrierung unter der Voraussetzung einer Makro-Mikro-Wissenslücke durchgeführt werden. 3)Kalibrierung Agenten-basierter Simulationen auf der Mikro-Ebene unter der Voraussetzung, dass zur Kalibrierung einzelner Agentenmodelle nicht ausreichend und potentiell verfälschte Daten zur Verhaltensvalidierung zur Verfügung stehen. Hierzu wird in dieser Arbeit das sogenannte Makro-Mikro-Verfahren zur Kalibrierung von Agenten-basierten Simulationen entwickelt. Das Verfahren besteht aus einem Basisverfahren, das im Verlauf der Arbeit um verschiedene Zusatzverfahren erweitert wird. Das Makro-Mikro-Verfahren und seine Erweiterungen sollen dazu dienen, die Modellkalibrierung trotz stark verrauschter Daten und eingeschränktem Wissen über die Wirkungszusammenhänge im Originalsystem geeignet zu ermöglichen und dabei den Kalibrierungsprozess zu beschleunigen: 1) Makro-Mikro-Kalibrierungsverfahren: Das in dieser Arbeit entwickelte Makro- Mikro-Verfahren unterstützt den Nutzer durch eine kombinierte Kalibrierung auf der Mikro- und der Makro-Beobachtungsebene, die gegebenenfalls durch Zwischenebenen erweitert werden kann. Der Grundgedanke des Verfahrens besteht darin, das Kalibrierungsproblem in eines auf aggregierter Verhaltensebene und eines auf der Ebene des Mikro-Agentenverhaltens aufzuteilen. Auf der Makro-Ebene wird nach validen idealen aggregierten Verhaltensmodellen (IVM) der Agenten gesucht. Auf der Mikro-Ebene wird versucht die individuellen Modelle der Agenten auf Basis des erwünschten Gesamtverhaltens und der ermittelten IVM so zu kalibrieren, das insgesamt Simulationsverhalten entsteht, das sowohl auf Mikro- als auch auf Makro-Ebene valide ist. 2) Erweiterung 1: Robuste Kalibrierung: Um den Umgang mit potentiell verrauschten Validierungskriterien (d.h. mit verrauschten Daten über ein Originalsystem, auf denen die Validierungskriterien der Simulation beruhen) und Modellteilen während der Kalibrierung von ABS zu ermöglichen, wird eine robuste Kalibrierungstechnik zur Anwendung im Makro-Mikro-Verfahren entwickelt. 3) Erweiterung 2: Kalibrierung mit Heterogenitätssuche: Als zweite Erweiterung des Makro-Mikro-Verfahrens wird ein Verfahren entwickelt, das das Problem des unklaren Detaillierungsgrades von ABS auf der Ebene der Parameterwerte adressiert. Prinzipiell kann zwar jeder Agent unterschiedliche Parameterwerte verwenden, obwohl eine geringere Heterogenität zur Erzeugung validen Verhaltens ausreichend wäre. Die entwickelte Erweiterung versucht, während der Kalibrierung, eine geeignete Heterogenitätsausprägung für die Parameterwerte der Agenten zu ermitteln. Unter einer Heterogenitätsausprägung wird dabei eine Einteilung der simulierten Agenten in Gruppen mit jeweils gleichen Parameterwerten verstanden. Die Heterogenitätssuche dient dazu, einen Kompromiss zu finden zwischen der Notwendigkeit, sehr große Parametersuchräume durchsuchen zu müssen und gleichzeitig den Suchraum so klein wie möglich halten zu wollen. / In this doctoral thesis the problem of calibrating agent-based simulations (ABS) is treated, i.e. the problem to adjust the parameter values of an agent-based simulation model to achieve valid simulation behavior. The calibration problem for simulations per se is not new and is an active part of research in the context of traditional simulation paradigms, such as the macro-simulation. Compared to the problems considered there the problems for ABS can be distinguished by several additional challenges that complicate the direct application of existing calibration procedures in a limited time, or challenges that do not allow applying existing procedures at all. The goal of this thesis is to assist the user in the calibration of ABS on the basis of incomplete and potentially noisy data or knowledge and in dealing with large amounts of parameter values if an ABS with many individual agents needs to be calibrated. The thesis covers the following three main topics: 1) Simplification of the calibration of many agent parameter values on the micro-level in ABS. This is done by exploiting the specific structure of ABS (i.e. that an ABS constitutes of a lattice of agent models). 2) Calibration of agent-based simulations, so that valid simulation behavior is created on all relevant behavior observation levels (at least micro- and macro-level). This needs to be possible without having full knowledge about how the macro observation level behavior constitutes from the modeled micro behavior. 3) Calibration of agent-based simulations on the micro-level under the constraint that only partial and potentially noisy data for testing and validation of single individual agent models is available. To achieve this the so-called “Macro-Micro Procedure” for calibrating agent-based simulations is developed. The approach consists of a basic procedure that is extended in the course of the work with various additional techniques: 1)Macro-Micro-Calibration Procedure: The Macro-Micro Procedure supports the user by applying a combined calibration on the micro and the macro-observation level, which can optionally be expanded using additional intermediate levels. The basic idea of the procedure consists of separating the calibration problem into one at the aggregate behavior level and one at the level of the micro-agent behavior. At the macro level, valid simulation behavior for ideal aggregate behavior models (IAM) of agents is being determined. At the micro level, the goal is to calibrate the models of the individual agents based on the desired overall behavior and the determined IAM from the macro level. Upon completion the simulation behavior created shall be valid both at the micro and also at a macro level. 2)Extension 1: Robust Calibration: In order to deal with potentially noisy validation criteria and model parts (i.e. with noisy data about the original system from which the validation criteria of the simulation are created) a robust calibration technique is developed that can be used as part of the Macro-Micro-Procedure. 3)Extension 2: Calibration with heterogeneity search: The second extension of the Macro-Micro-Procedure addresses the problem of an unclear level of detail on the level of the parameter values. Theoretically it is possible to use different parameter values for each individual simulated agent which leads to a huge parameter search space. Often it is however sufficient to use a lower heterogeneity in the parameter values to generate valid behavior which would allow calibration in a smaller search space. The developed extension attempts to determine such a suitable heterogeneity manifestation for the parameter values of the agents as part of the calibration process itself. A heterogeneity manifestation is performed by dividing the agents into groups of agents with homogenous parameter values. The developed heterogeneity search offers a compromise between the necessity of having to search very large parameter search spaces and the goal to keep the search space as small as possible during the calibration.

Machine condition monitoring is gaining importance in industry due to the need to increase machine reliability and decrease the possible loss of production due to machine breakdown. Often the data available to build a condition monitoring system does not fully represent the system. It is also often common that the data becomes available in small batches over a period of time. Hence, it is important to build a system that is able to accommodate new data as it becomes available without compromising the performance of the previously learned data. In real-world applications, more than one condition monitoring technology is used to monitor the condition of a machine. This leads to large amounts of data, which require a highly skilled diagnostic specialist to analyze. In this thesis, artificial intelligence (AI) techniques are used to build a condition monitoring system that has incremental learning capabilities. Two incremental learning algorithms are implemented, the first method uses Fuzzy ARTMAP (FAM) algorithm and the second uses Learn++ algorithm. In addition, intelligent agents and multi-agent systems are used to build a condition monitoring system that is able to accommodate various analysis techniques. Experimentation was performed on two sets of condition monitoring data; the dissolved gas analysis (DGA) data obtained from high voltage bushings and the vibration data obtained from motor bearing. Results show that both Learn++ and FAM are able to accommodate new data without compromising the performance of classifiers on previously learned information. Results also show that intelligent agent and multi-agent system are able to achieve modularity and flexibility.

In real-world conversational groups, interactants adjust their body position and orientation relative to one another in order to see and hear clearly. We use an agent-based modelling approach to compare alternative models for simulating the spatial patterns of conversational groups. The models are based on simple rules that control the movement, positioning, and orientation behaviour of individual agents, which in turn leads to the emergence of agent clusters. We identify which model alternative produces agent clusters with characteristics typical of real-world conversational groups. The centroid-based approach, where agents readjust their position and orientation with respect to the group centroid point, is a commonly used method to simulate conversational groups, but has not been empirically validated. This thesis replicates, evaluates, and validates the centroid-based model in a systematic way. Another model, where agents perform positional-orientational readjustments to see as many neighbours as possible within a 180 field of view, called the field-of-view approach is proposed, implemented, evaluated, and validated. Analysis of the spatial patterns of conversational groups has hitherto mostly relied on visual verification. We, novelly, use a combination of qualitative and quantitative methods to analyse the spatial patterns of conversational groups. Evaluations show that the field of- view model and centroid-based model produce agent clusters with significantly different social, spatial, and temporal characteristics. Validation is performed using a dataset which captures the spatial behaviour of 21 participants for the entire duration of a party. This validation shows that the characteristics of agent clusters resulting from the field-of-view model most closely reflects the characteristics of real-world conversational groups. We also show that a local neighbourhood influence works better than an extended neighbourhood influence to simulate conversational groups. The influence of objects in the environment on the spatial patterns of agent clusters are also discussed.