M.Ing. / The Graphics Processor Unit (GPU) has been in the gaming industry for several years now. Of late though programmers and scientists have started to use the parallel processing or stream processing capabilities of the GPU in general numerical applications. The Monte Carlo method is a processing intensive methods, as it evaluates systems with stochastic components. The stochastic components require several iterations of the systems to develop an idea of how the systems reacts to the stochastic inputs. The stream processing capabilities of GPUs are used for the analysis of such systems. Evaluating low-cost Inertial Measurement Units (IMU) for utilisation in Inertial Navigation Systems (INS) is a processing intensive process. The non-deterministic or stochastic error components of the IMUs output signal requires multiple simulation runs to properly evaluate the IMUs performance when applied as input to an INS. The GPU makes use of stream processing, which allows simultaneous execution of the same algorithm on multiple data sets. Accordingly Monte Carlo techniques are applied to create trajectories for multiple possible outputs of the INS based on stochastically varying inputs from the IMU. The processing power of the GPU allows simultaneous Monte Carlo analysis of several IMUs. Each IMU requires a sensor error model, which entails calibration of each IMU to obtain numerical values for the main error sources of lowcost IMUs namely scale factor, non-orthogonality, bias, random walk and white noise. Three low-cost MEMS IMUs was calibrated to obtain numerical values for their sensor error models. Simultaneous Monte Carlo analysis of each of the IMUs is then done on the GPU with a resulting circular error probability plot. The circular error probability indicates the accuracy and precision of each IMU relative to a reference trajectory and the other IMUs trajectories. Results obtained indicate the GPU to be an alternative processing platform, for large amounts of data, to that of the CPU. Monte Carlo simulations on the GPU was performed 200 % faster than Monte Carlo simulations on the CPU. Results obtained from the Monte Carlo simulations, indicated the Random Walk error to be the main source of error in low-cost IMUs. The CEP results was used to determine the e ect of the various error sources on the INS output.

Orientador: Paulo Roberto Gardel Kurka / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Mecânica / Made available in DSpace on 2018-08-27T20:14:23Z (GMT). No. of bitstreams: 1 Silva_PedroRamondeMello_M.pdf: 5875188 bytes, checksum: c2d175b05fd3a31dbc7149f0b1b417b1 (MD5) Previous issue date: 2015 / Resumo: A navegação autônoma de robôs tem sido alvo de pesquisa. Um dos problemas a ser resolvido em navegação é a localização de um robô móvel enquanto este se locomove, ou seja, estimar o deslocamento e orientação para calcular a trajetória realizada. Um sistema de navegação inercial é um método de realizar a localização do robô utilizando sensores inerciais (acelerômetro, giroscópio e magnetômetro). Este trabalho apresenta a utilização de um dispositivo móvel baseado no sistema operacional Android, como UMI (Unidade de Medida Inercial), para um desenvolvimento de um sistema de navegação inercial de um robô terrestre. Para o calculo de orientação, foi aplicado o filtro complementar para melhorar o desempenho da estimação de orientação calculada pelo acelerômetro/magnetômetro combinado com o giroscópio. Para estimação de deslocamento, aplicou-se integral dupla nos dados de aceleração e propõe-se uma metodologia de movimento modulado para minimizar os erros de posição e desvios no sinal de velocidade devido ao processo de integração numérica e erros de devido a componente de corrente continua do acelerômetro / Abstract: The autonomous navigation of mobile robots have been the subject of many research. One of the problems to be solved in navigation, is the localization of a mobile robot as it moves around, in other words, to estimate the displacement and orientation to calculate the trajectory performed. An inertial navigation system is a method of performing the localization of the robot using inertial sensors (accelerometer, gyroscope and magnetometer). This work presents the use of a mobile device based on the Android platform, as IMU (Inertial Measurement Unit), for the development of an inertial navigation system of a robot. For guidance calculation, the complementary filter has been applied to improve the performance of guidance computed by accelerometer / magnetometer combined with the gyro. For displacement estimation, a double integral in the acceleration data was applied. Furthermore, a motion modulated methodology is proposed to minimize position errors and velocity drift due to numerical integration process and accelerometer offset errors / Mestrado / Mecanica dos Sólidos e Projeto Mecanico / Mestre em Engenharia Mecânica

The aim of this work was to create a self-localization unit. The basis of the unit is an electronic platform Netduino Plus, which is connected with a measurement unit MTi-G and a camera based sensor, which is able to detect the position of points on the schools testing polygon. Configuration of the measurement unit MTi-G was done by a modified previously developed program. The program of the platform uses lover level communication with the sensor without using predefined classes from the manufacturer. This should improve the speed of processing the data from the sensor. To check the accuracy of the position provided by the self-localization unit another program was created. This program receives messages form a more accurate localization system that is mounted to the robot. The localization system is based on Trimble DB982 receiving modules.

No description available.

The aim of this thesis is to advance the control and estimation schemes for multirotors, and more specifically the Aerial Robotic Worker, in order to progress towards the necessary control and estimation performance for robust control, cooperation and collaboration. Towards this envisioned aim, this Licentiate thesis will present the following main research contributions: a) a singularity-free attitude controller for the attitude problem has been established, that does not have the inherent drawbacks of Euler angle or Direction Cosine Matrix based approaches, b) a generalized estimation scheme for attitude, position and parameter estimation will be presented that has the merit of low computational footprint, while it is robust towards magnetic disturbances and able to identify key parameters in the model of an Aerial Robotic Worker, c) an method for estimating the induced vibration frequencies on the multirotor’s frame, and the respective amplitudes, that relies on notch filtering for attenuating the induced vibrations, and d) a theoretical establishment, as well as an experimental development and evaluation of a variable pitch propeller model to add additional degrees of freedom and increase the robustness of an Aerial Robotic Worker. In the first part of this thesis the main contributions of the previous research approaches will be highlighted, while in the second part of the thesis the corresponding and in full detail articles will be presented.

No description available.

No description available.

No description available.

No description available.

No description available.