The ePUMA architecture is a novel master-multi-SIMD DSP platform aimed at low-power computing, like for embedded or hand-held devices for example. It is both a configurable and scalable platform, designed for multimedia and communications. Numbers with both integer and fractional parts are often used in computers because many important algorithms make use of them, like signal and image processing for example. A good way of representing these types of numbers is with a floating-point representation. The ePUMA platform currently supports a fixed-point representation, so the goal of this thesis will be to implement twelve basic floating-point arithmetic operations and two conversion operations onto an already existing datapath, conforming as much as possible to the IEEE 754-2008 standard for floating-point representation. The implementation should be done at a low hardware and power consumption cost. The target frequency will be 500MHz. The implementation will be compared with dedicated DesignWare components and the implementation will also be compared with floating-point done in software in ePUMA. This thesis presents a solution that on average increases the VPE datapath hardware cost by 15% and the power consumption increases by 15% on average. Highest clock frequency with the solution is 473MHz. The target clock frequency of 500MHz is thus not achieved but considering the lack of register retiming in the synthesis step, 500MHz can most likely be reached with this design.

Improved actuator point-to-point positioning performance, as measured by settling time, has been demonstrated in the context of manufacturing automation applications such as circuit board assembly and other product-transfer operations. The control objective is to move a single mass in a single axis from a starting position to a target position following the fastest possible motion trajectory while meeting final-position accuracy requirements. The actuator's achievable force that is available for acceleration is the fundamental variable that determines optimal settling time. The actuator technology employed is the linear variable reluctance motor. Mathematical motor models and simulation programs have been developed to perform several tasks necessary for demonstrating improved actuator performance: (i) optimal commutation under force ripple constraints has been computed to determine ripple-specified force limits and to provide excitation waveforms necessary for force production, (ii) motion profiles for several positioning task scenarios have been generated based on computed ripple-specified force limits, (iii) state space integral position control simulations have been performed to evaluate the degree of success of the proposed relaxation of force ripple constraints in improving settling time and (iv) the computed settling times for positioning tasks have been examined in relation to the copper losses associated with them in order to assess the trade-off. It has been shown that higher force capability is achieved when force-ripple constraints, which have been customarily emphasized in positioning applications, are relaxed. The higher capability is exploited by adopting faster motion trajectories, which are then imposed under feedback control to achieve faster settling time. Improved force capability with relaxed ripple constraints is demonstrated by generating average force versus speed capability curves under ripple constraints ranging from minimal ripple to unconstrained ripple. Improved positioning performance, with relaxed ripple constraints and without violating the final-position accuracy specification, is demonstrated by computing and comparing settling time for multiple positioning tasks with trajectories based on both extremes of force capability, lowest (no-ripple) and highest (unconstrained-ripple) force limits. The results have been demonstrated for two LVR motor configurations: one motor configuration represents typical (switched) linear and rotary variable reluctance motors while the other exhibits features of both switched and synchronous varieties of variable reluctance motors.

No description available.

No description available.

No description available.

No description available.

Europa-Univ. Viadrina, Diss.--Frankfurt/Oder, 2002.

Thesis (M. Sc.)--University of Hong Kong, 1992.

Zugl.: Münster (Westfalen), Universiẗat, Diss., 1990.

A utilização do sistema LiDAR na obtenção de dados da superfície da Terra vem se disseminando cada vez mais pelo alto desempenho na aquisição da informação e pela efetiva utilização dos produtos e subprodutos. Diversos segmentos passaram a adotar os produtos LiDAR como insumo básico e fundamental em suas rotinas de trabalho e estudos. O sucesso em um projeto que envolve aquisição e utilização desse tipo de dado está atrelado diretamente à definição e seu planejamento. O usuário deve ser capaz de definir o escopo básico e as diretrizes técnicas fundamentais que irão garantir que a demanda final seja alcançada. Diante disto se faz necessário elaborar um planejamento estabelecendo a melhor configuração para a aquisição dos dados e também os tipos de informações que serão extraídas destes produtos. Referente ao primeiro aspecto pode-se dizer que este é a base para a todo o projeto. Por meio dele é garantido a obtenção de produtos conforme a necessidade do usuário (resoluções espaciais dos produtos, o nível de detalhamento dos objetos, representação da topografia e outros). Muitos dos usuários que contratam serviços de perfilamento LiDAR não possuem embasamento técnico suficiente para definir a melhor especificação a ser adotada. Isto faz com que a maioria deles opte por adquirir uma alta densidade de pontos que, muitas vezes é desnecessária e ainda que atendam à demanda final tornam o projeto financeiramente oneroso. Esta pesquisa mostra que para um ambiente urbano regular, nuvens de pontos com baixas densidades (4 pts/m² e 8 pts/m²) apresentam uma equivalência na qualidade geométrica para produtos e subprodutos obtidos que serão utilizados para determinadas aplicações, não havendo a necessidade de utilizar nuvens com uma alta densidade (16 pts/m²) em projetos que utilizam estes dados em estudos altimétricos: geração do Modelo Digital de Terreno, curvas de nível, pontos cotados e também planimétricos: Modelo Digital de Elevação (Superfície e Normalizada) e seus derivados (altura e contorno de objetos, imagem de intensidade, cobertura vegetal e outros). / The use of the LiDAR system in obtaining data from the Earth\'s surface has been increasingly disseminated by the high performance in the acquisition of information and the effective use of products and by-products. Several segments started to adopt LiDAR products as basic and fundamental input in their works and studies. Success in a project that involves the acquisition and use of this type of data is directly linked to the definition and its planning. The user must be able to define the basic scope and the fundamental technical guidelines that will guarantee that the final demand is reached . In view of this, it is necessary to prepare a planning establishing the best configuration for the data acquisition and also the types of information that will be extracted from these products. Regarding the first aspect can be said that this is the basis for the whole project. Through it is guaranteed to obtain products according to the user is needs (spatial resolutions of the products, the level of detail of the objects, representation of the topography and others). Many of the users who hire LiDAR profiling services do not have sufficient technical background to define the best specification to adopt. This makes the majority of them choose to acquire a high density of points that is often unnecessary and even if they meet the final demand make the project financially costly. This research shows that for a regular urban environment, low point density clouds (4 pts/m² and 8 pts/m²) present an equivalence in the geometric quality for products and by-products obtained that will be used for certain applications. The need to use clouds with a high density (16 pts/m²) in projects that use this data in altimetric studies: generation of Digital Terrain Model, contourlines, quoted points and also planimetric: Digital Elevation Model (Surface and Normalized) and its derivatives (height and contour of objects, intensity image, vegetation cover and others).