During the past decade, a new class of knowledge representation has emerged known as structured distributed representation (SDR). A number of schemes for encoding and manipulating such representations have been developed; e. g. Pollack's Recursive Auto-Associative Memory (RAAM), Kanerva's Binary Spatter Code (BSC), Gayler's MAP encoding, and Plate's Holographically Reduced Representations (HRR). All such schemes encode structural information throughout the elements of high dimensional vectors, and are manipulated with rudimentary algebraic operations. <br /><br /> Most SDRs are very compact; components and compositions of components are all represented as fixed-width vectors. However, such compact compositions are unavoidably noisy. As a result, resolving constituent components requires a cleanup memory. In its simplest form, cleanup is performed with a list of vectors that are sequentially compared using a similarity metric. The closest match is deemed the cleaned codevector. <br /><br /> While SDR schemes were originally designed to perform cognitive tasks, none of them have been demonstrated in a neurobiologically plausible substrate. Potentially, mathematically proven properties of these systems may not be neurally realistic. Using Eliasmith and Anderson's (2003) Neural Engineering Framework, I construct various spiking neural networks to simulate a general cleanup memory that is suitable for many schemes. <br /><br /> Importantly, previous work has not taken advantage of parallelization or the high-dimensional properties of neural networks. Nor have they considered the effect of noise within these systems. As well, additional improvements to the cleanup operation may be possible by more efficiently structuring the memory itself. In this thesis I address these lacuna, provide an analysis of systems accuracy, capacity, scalability, and robustness to noise, and explore ways to improve the search efficiency.

Modelling mechanical systems using symbolic equations can provide many advantages over the more widely-used numerical methods of modelling these systems. The use of symbolic equations produces more efficient models, which can be used for many purposes such as real-time simulation and control. However, the number, complexity, and computational efficiency of these equations is highly dependent on which coordinate set was used to model the system. One method of modelling a mechanism's topology and formulating its symbolic equations is to model the system using a graph-theoretical approach. This approach models mechanisms using a linear graph, from which spanning trees can be used to define a mechanism's coordinate set. This report develops two tree selection algorithms capable of estimating the tree set, and hence coordinate set, that produces models having the fastest forward dynamic simulation times. The first tree selection algorithm is a heuristic-based algorithm that tries to find the coordinate set containing the minimal possible number of modelling variables. Most of this algorithm's heuristics are based on tree selection criteria found in the literature and on observations of a series of benchmark problems. It uses the topology information provided by a system's graph to find the coordinates set for the given system that produce very low simulation times of the system. The second tree selection algorithm developed in this report also uses graph theory. It bases most of its heuristics on observations of one of the methods developed to obtain a mechanical system's symbolic equations using graph theory. This second algorithm also makes use of, and improves upon, a few of the heuristics developed in the first tree selection algorithm. A series of examples for both algorithms will demonstrate the computational efficiency obtained by using the modelling variables found by the automated tree selection algorithms that are proposed in this report.

Multiple criteria decision analysis (MCDA) techniques are developed to address challenging classification problems arising in engineering management and elsewhere. MCDA consists of a set of principles and tools to assist a decision maker (DM) to solve a decision problem with a finite set of alternatives compared according to two or more criteria, which are usually conflicting. The three types of classification problems to which original research contributions are made are <ol> <li>Screening: Reduce a large set of alternatives to a smaller set that most likely contains the best choice. </li> <li>Sorting: Arrange the alternatives into a few groups in preference order, so that the DM can manage them more effectively. </li> <li>Nominal classification: Assign alternatives to nominal groups structured by the DM, so that the number of groups, and the characteristics of each group, seem appropriate to the DM. </ol> Research on screening is divided into two parts: the design of a sequential screening procedure that is then applied to water resource planning in the Region of Waterloo, Ontario, Canada; and the development of a case-based distance method for screening that is then demonstrated using a numerical example. <br /><br /> Sorting problems are studied extensively under three headings. Case-based distance sorting is carried out with Model I, which is optimized for use with cardinal criteria only, and Model II, which is designed for both cardinal and ordinal criteria; both sorting approaches are applied to a case study in Canadian municipal water usage analysis. Sorting in inventory management is studied using a case-based distance method designed for multiple criteria ABC analysis, and then applied to a case study involving hospital inventory management. Finally sorting is applied to bilateral negotiation using a case-based distance model to assist negotiators that is then demonstrated on a negotiation regarding the supply of bicycle components. <br /><br /> A new kind of decision analysis problem, called multiple criteria nominal classification (MCNC), is addressed. Traditional classification methods in MCDA focus on sorting alternatives into groups ordered by preference. MCNC is the classification of alternatives into nominal groups, structured by the DM, who specifies multiple characteristics for each group. The features, definitions and structures of MCNC are presented, emphasizing criterion and alternative flexibility. An analysis procedure is proposed to solve MCNC problems systematically and applied to a water resources planning problem.

published_or_final_version / Electrical and Electronic Engineering / Doctoral / Doctor of Philosophy

Not available / text

Not available / text

Not available / text

published_or_final_version / abstract / toc / Industrial and Manufacturing Systems Engineering / Master / Master of Philosophy

No description available.

A free-ranging automated guided vehicle incorporating navigation and radio communication for use in a fully automated flexible manufacturing system has been developed. A vehicle, operating as a complete subsystem, was built and tested in an integrated control environment and proved to have promising results. various radio communication techniques are examined and the design and testing of a low cost, wireless, two way communication link is detailed. A novel, flexible infrared navigation technique was developed and incorporated into the AGV subsystem. Path planning and a flexible real time path modification system was formulated using an innovative program with an interpolative visual display unit and digitiser. Data transfer to and from the vehicles in a real time integrated system is covered. System integration for an free-ranging automatic guided vehicle is discussed covering aspects of communication, mapping and navigation. Specific needs for a free-ranging automatic guided vehicle, are presented. The unique design features of navigation and mapping outlined in this thesis has resulted in a low cost, free-ranging, autonomous automatic guided vehicle. / Thesis (Ph.D.)-University of Natal, Durban, 1992.