In the thesis, we present the structures of interacting multiple model (IMM) algorithm-based filter design for real-time motion orientation estimation and tracking by using inertial sensor measurements in three-dimensional space. The major sensor such as gyroscope, though has high-sensitivity characteristics, suffers from bias build-up and error drift over time. The complementary sensors such as accelerometer and magnetometer, on the other hand, have low sensitivity, but do not suffer from bias problems. By using individual inertial and magnetic sensors, measurements of multiple modes can be interactively computed. The IMM based designs show the advantages of weighting individual sensors in different motion states. We propose a signal processing architecture based on the IMM algorithm. It is composed of three parallel Kalman filters (KFs), each deals with measured signals from accelerometer, magnetometer and gyroscope, respectively. The accelerometer cannot effectively sense the rotation around the vertical axis; while the magnetometer can only sense the rotation around vertical axis. Therefore, estimation accuracy with the parallel filtering arrangement of the IMM algorithm-based structure may be affected. A scheme using the residual signal, which is computed in the IMM, provides the information of gyroscope-based KF to the other two filters for feasible calculation of update weights. Related research also usually combined the information of major and complementary sensors in estimator designs. In the literature, existing ¡§Triad¡¨ methods with quaternion-based extended Kalman filter (EKF), process the measurements from major and complementary sensors. To compensate the functions, we propose to use a gyroscope-based EKF and a Triad EKF in forming a parallel multiple model-based structure. The analysis and performance evaluation shows advantages and disadvantages of using EKFs and KFs in IMM-based filtering approachs. Simulation results validate the proposed estimator design concept, and show that the scheme is capable of reducing the overall estimation errors by flexible computation of model weights.

Recording a video sequence with a camera during movement often produces blurred results. This is mainly due to motion blur which is caused by rapid movement of objects in the scene or the camera during recording. By correcting for changes in the orientation of the camera, caused by e.g. uneven terrain, it is possible to minimize the motion blur and thus, produce a stabilized video. In order to do this, data gathered from a gyroscope and the camera itself can be used to measure the orientation of the camera. The raw data needs to be processed, synchronized and filtered to produce a robust estimate of the orientation. This estimate can then be used as input to some automatic control system in order to correct for changes in the orientation This thesis focuses on examining the possibility of such a stabilization. The actual stabilization is left for future work. An evaluation of the hardware as well as the implemented methods are done with emphasis on speed, which is crucial in real time computing. / En videosekvens som spelas in under rörelse blir suddig. Detta beror främst på rörelseoskärpa i bildrutorna orsakade av snabb rörelse av objekt i scenen eller av kameran själv. Genom att kompensera för ändringar i kamerans orientering, orsakade av t.ex. ojämn terräng, är det möjligt att minimera rörelseoskärpan och på så sätt stabilisera videon. För att åstadkomma detta används data från ett gyroskop och kameran i sig för att skatta kamerans orientering. Den insamlade datan behandlas, synkroniseras och filtreras för att få en robust skattning av orienteringen. Denna orientering kan sedan användas som insignal till ett reglersystem för att kompensera för ändringar i kamerans orientering. Denna avhandling undersöker möjligheten för en sådan stabilisering. Den faktiska stabiliseringen lämnas till framtida arbete. Hårdvaran och de implementerade metoderna utvärderas med fokus på beräkningshastighet, som är kritiskt inom realtidssystem.

The purpose of this thesis is to study and investigate a practical and efficient implementation of corner orientation detection using multisteerable filters. First, practical theory involved in applying multisteerable filters for corner orientation estimation is presented. Methods to improve the efficiency with which multisteerable corner filters are applied to images are investigated and presented. Prior research in this area presented an optimization equation for determining the best match of corner orientations in images; however, little research has been done on optimization techniques to solve this equation. Optimization techniques to find the maximum response of a similarity function to determine how similar a corner feature is to a multioriented corner template are also explored and compared in this research.

No description available.

Today Image acquisition equipment produces huge amounts of data that needs to be processed. Often the data describes signals with a dimensionality higher then 2, as with ordinary images. This introduce a problem when it comes to process this high dimensional data since ordinary signal processing tools are no longer suitable. New faster and more efficient tools need to be developed to fully exploit the advantages with e. g. a 3D CT-scan. One such tool is filternets, a layered networklike structure, which the signal propagates through. A filternet has three fundamental advantages which will decrease the filtering time. The network structure allows complex filter to be decomposed into simpler ones, intermediate result may be reused and filters may be implemented with very few nonzero coefficients (sparse filters). The aim of this study has been to create an implementation for filternets and optimize it with respect to execution time. Specially the possibility to use filternets that approximates a harmonic filterset for estimating orientation in 3D signals is investigated. Tests show that this method is up to about 30 times faster than a full filterset consisting of dense filters. They also show a slightly larger error in the estimated orientation compared with the dense filters, this error should however not limit the usability of the method.

This thesis is concerned with attitude estimation of small flying robots using low cost, small-sized inertial and magnetic sensors. A quaternion-based extended Kalman filter is used, which adaptively compensates external acceleration. External acceleration is the main source of estimation error. If external acceleration is detected, the accelerometer measurement covariance matrix of the Kalman filter is adjusted. The proposed algorithms are verified through experiments. Selected algorithm is implemented on STM32F4DISCOVERY development board.

京都大学 / 0048 / 新制・課程博士 / 博士(農学) / 甲第22783号 / 農博第2426号 / 新制||農||1081(附属図書館) / 学位論文||R2||N5303(農学部図書室) / 京都大学大学院農学研究科地域環境科学専攻 / (主査)教授 近藤 直, 教授 飯田 訓久, 准教授 小川 雄一 / 学位規則第4条第1項該当 / Doctor of Agricultural Science / Kyoto University / DFAM

This thesis develops an alternative sensor fusion approach for object orientation using low-cost MEMS inertial sensors. The alternative approach focuses on the unique challenges of small UAVs. Such challenges include the vibrational induced noise onto the accelerometer and bias offset errors of the rate gyroscope. To overcome these challenges, a sensor fusion algorithm combines the measured data from the accelerometer and rate gyroscope to achieve a single output free from vibrational noise and bias offset errors. One of the most prevalent sensor fusion algorithms used for orientation estimation is the Extended Kalman filter (EKF). The EKF filter performs the fusion process by first creating the process model using the nonlinear equations of motion and then establishing a measurement model. With the process and measurement models established, the filter operates by propagating the mean and covariance of the states through time. The success of EKF relies on the ability to establish a representative process and measurement model of the system. In most applications, the EKF measurement model utilizes the accelerometer and GPS-derived accelerations to determine an estimate of the orientation. However, if the GPS-derived accelerations are not available then the measurement model becomes less reliable when subjected to harsh vibrational environments. This situation led to the alternative approach, which focuses on the correlation between the rate gyroscope and accelerometer-derived angle. The correlation between the two sensors then determines how much the algorithm will use one sensor over the other. The result is a measurement that does not suffer from the vibrational noise or from bias offset errors.

Orientation estimation is a very well known topic in many fields such as in aerospace or robotics. However, the sensors used are usually very ex- pensive, heavy and big, which make them not suitable for IoT (Internet of Things) based applications. This thesis presents a study of how different sensor fusion algorithms perform in low cost hardware and in high acceler- ation scenarios. For this purpose, an Arduino MKR1000 is used together with an accelerometer, gyroscope and magnetometer. The objective of the thesis is to choose the most suitable algorithm for the purposed practical application, which consists on attaching the device to a moving object, such as a skate board or a bike. Once the orientation is estimated, a movement recognition algorithm that was developed is able to match what trick or movement was performed. The algorithm chosen was the Madgwick one with some minor adjustments, which uses quaternions for the estimation and is very resilient when the device is under strong external accelerations.

Detta examensarbete har genomförts med syftet att utveckla Inno-x företagets system, som är avsett för vardagsidrottare för att mäta neuromuskulära aktiviteter i underkroppen med hjälp av modern teknologi. Systemet omfattar en tröghetsmätningsenhet (IMU) med accelerometer, gyroskop och en EMG-sensor (elektromyografi). Denna konfiguration möjliggör noggrann övervakning av neuromuskulära aktiviteter genom analys av svar på träning. Studiens mål var att identifiera en effektiv sensor för mätning av horisontella hoppavstånd och att utveckla en algoritm som sedan ska integreras i företagets produkt. Produkten kommer att använda magnetometer och IMU för att tolka mänskliga rörelser och för att förbättra noggrannheten i företagets mätningssystem. Processen inkluderar förbättring av mätningarnas noggrannhet, integration av teknik med biomekaniska principer, utvärdering av kalibreringstekniker för magnetometeravläsningar, kombination av sensorer för rörelseanalys och genomförande av utvärdering med olika åldersgrupper som består av 10 deltagare för att bedöma systemets effektivitet. Även om ingen av metoderna helt uppnådde den önskade noggrannheten inom ±5 cm, visade alla metoder god prestanda för olika tillämpningar. Detta antyder att implementeringen av en kalibrerad magnetometer potentiellt kan förbättra systemets noggrannhet vid bestämning av horisontella hoppavstånd, dock endast med en liten marginal, eftersom studien visade att med kalibrerade magnetometer RMSE (Root Mean Square Error) ökat med 0.99 cm. Ytterligare forskning rekommenderas för att undersöka nya sätt att kalibrera sensorer och integrera dem för mer precisa avläsningar. Dock bör det beaktas att magnetometeravläsningar påverkas av miljöfaktorer. Dessutom är det viktigt att skapa ett användarvänligt gränssnitt som gör det möjligt för idrottare att enkelt spåra och analysera sina prestandadata. / This thesis has been conducted with the objective of developing the Inno-X company's system, which is intended for everyday athletes to measure neuromuscular activities in the lower body using modern technology. The system includes an Inertial Measurement Unit (IMU) with an accelerometer, gyroscope, and an Electromyography (EMG) sensor. This configuration enables accurate monitoring of neuromuscular activities through the analysis of responses to training. The study's goal was to identify an effective sensor for measuring horizontal jump distances and to develop an algorithm that would then be integrated into the company's product. The product will use a magnetometer and IMU to interpret human movements and to improve the accuracy of the company's measurement system. The process includes improving the accuracy of measurements, integrating technology with biomechanical principles, evaluating calibration techniques for magnetometer readings, combining sensors for motion analysis, and conducting evaluations with different age groups consisting of 10 participants to assess the system's effectiveness. Although none of the methods fully achieved the desired accuracy within ±5 cm, all methods showed good performance for various applications. This suggests that the implementation of a calibrated magnetometer could potentially improve the system's accuracy in determining horizontal jump distances, albeit only by a small margin, as the study showed that with calibrated magnetometers, the Root Mean Square Error (RMSE) increased by 0.99 cm. Further research is recommended to explore new ways to calibrate sensors and integrate them for more precise readings. However, it should be considered that magnetometer readings are affected by environmental factors. Additionally, it is important to create a user-friendly interface that enables athletes to easily track and analyze their performance data.