Human Activity Recognition (HAR) is a growing field of research concerned with classifying human activities from sensor data. Modern smartphones contain numerous sensors that could be used to identify the physical activities of the smartphone wearer, which could have applications in sectors such as healthcare, eldercare, and fitness. This project aims to use smartphone sensor data together with machine learning to perform HAR on the following human locomotion activities: standing, walking, running, ascending stairs, descending stairs, and biking. The classification was done using a random forest classifier. Furthermore, in the special case of walking, an algorithm that can count the number of steps in a given data sequence was developed. The step counting algorithm was not based on a previous implementation and could therefore be considered novel. The step counter achieved a testing accuracy of 99.1\% and the HAR classifier a testing accuracy of 100\%. It is speculated that the abnormally high accuracies can be attributed primarily to the lack of data diversity, as in both cases only two persons collected the data. / Mänsklig aktivitetsigenkänning är ett växande forskningsområde som handlar om att klassificera mänskliga aktiviteter från sensordata. Moderna mobiltelefoner innehåller många sensorer som kan användas för att identifiera de fysiska aktiviteterna som bäraren utför, vilket har tillämpningar inom sektorer som sjukvård, äldreomsorg och personlig hälsa. Detta projekt använder sensordata från mobiltelefoner tillsammans med maskininlärning för att utföra aktivitetsigenkänning på följande aktiviteter: stå, gå, springa, gå uppför trappor, gå nedför trappor och cykla. Klassificeringen gjordes med hjälp av en ``random forest''-klassificerare. Vidare utvecklades en algoritm som kan räkna antalet steg i en given datasekvens som samlats in när användaren går. Stegräkningsalgoritmen baserades inte på en tidigare implementering och kan därför betraktas som ny. Stegräknaren uppnådde en testnoggrannhet på 99,1\% och aktivitetsigenkänningen en testnoggrannhet på 100\%. De oväntat höga noggrannheterna antas främst bero på bristen av diversitet i datan, eftersom den endast samlades in av två personer i båda fallen. / Kandidatexjobb i elektroteknik 2022, KTH, Stockholm

Human Activity Recognition (HAR) is a widelyresearched field that has gained importance due to recentadvancements in sensor technology and machine learning. InHAR, sensors are used to identify the activity that a person is performing.In this project, the six everyday life activities walking,biking, sitting, standing, ascending stairs and descending stairsare classified using smartphone accelerometer and gyroscope datacollected by three subjects in their everyday life. To performthe classification, two different machine learning algorithms,Artificial Neural Network (ANN) and Support Vector Machine(SVM) are implemented and compared. Moreover, we comparethe accuracy of the two sensors, both individually and combined.Our results show that the accuracy is higher using only theaccelerometer data compared to using only the gyroscope data.For the accelerometer data, the accuracy is greater than 95%for both algorithms and only between 83-93% using gyroscopedata. Also, there is a small synergy effect when using both sensors,yielding higher accuracy than for any individual sensor data, andreaching 98.5% using ANN. Furthermore, for all sensor types, theANN outperforms the SVM algorithm, having a greater accuracyby more than 1.5-9 percentage points. / Aktivitetsigenkänning är ett noga studeratforskningsområde som växt i popularitet på senare tid på grundav nya framsteg inom sensorteknologi and maskininlärning. Inomaktivitetsigenkänning använder man sensorer för att identifieravilken aktivitet en person utför. I det här projektet undersökervi de sex olika vardagsmotionsaktiviteterna gå, cykla, sitta, stå och gå i trappor (up/ner) med hjälp av data från accelerometeroch gyroskop i en smartphone som samlats in av tre olikapersoner. Två olika maskininlärningsalgoritmer implementerasoch jämförs: Artificial Neural Network (ANN) och SupportVector Machine (SVM). Vidare jämför vi noggranheten förde två sensorna, både individuellt och gemensamt. Våra resultvisar att noggranheten är större när enbart accelerometerdatananvänds jämfört med att använda enbart gyroskopdatan. Föraccelerometerdatan erhålls en noggranhet större än 95 % förbåda algoritmerna medan den siffran bara är mellan 83-93 %för gyroskopdatan. Dessutom existerar det en synergieffekt vidanvändande av båda sensorerna, och noggranheten når då 98.5% vid användande av ANN. Vidare visar våra resultat att ANNhar en noggranhet som är 1.5-9 procentenheter bättre än SVMför alla sensorer. / Kandidatexjobb i elektroteknik 2021, KTH, Stockholm

Gyroscopes are inertial sensors that measure the rate or angle of rotation. One of the most promising technologies for reaching a high-performance MEMS gyroscope has been development of the micro-hemispherical shell resonator. (μHSR) This thesis presents the electronic control and read-out interface that has been developed to turn the μHSR into a fully functional micro-hemispherical resonating gyroscope (μHRG) capable of measuring the rate of rotation. First, the μHSR was characterized, which both enabled the design of the interface and led to new insights into the linearity and feed-through characteristics of the μHSR. Then a detailed analysis of the rate mode interface including calculations and simulations was performed. This interface was then implemented on custom printed circuit boards for both the analog front-end and analog back-end, along with a custom on-board vacuum chamber and chassis to house the μHSR and interface electronics. Finally the performance of the rate mode gyroscope interface was characterized, showing a linear scale factor of 8.57 mv/deg/s, an angle random walk (ARW) of 34 deg/sqrt(hr) and a bias instability of 330 deg/hr.

We consider two models which lead to the prediction of a weight change of gyroscopes depending on the rate of rotation: mass-energy equivalence and Weber's force for gravitation. We calculate the order of magnitude of this effect in both models and show that Weber's model predicts a weight change depending on the spin axis orientation resembling close similarities to observed Earth flyby anomalies. however, our predicted effect is much smaller than the observed effect, which could explain why flyby anomalies were not detected anymore in recent spracecraft trajectories.

Esta dissertação apresenta a modelagem matemática de um giroscópio de controle de momento (CMG) e aplica técnicas de controle digital para o controle de atitude dessa planta. CMGs são atuadores importantes para o controle de atitude de corpos no espaço, tais como satélites e veículos espaciais. Desenvolve-se a modelagem da planta por meio da mecânica Lagrangiana e estudam-se técnicas específicas de controle PID (proporcional-integral-derivativo) e LQR (linear-quadratic regulator ) digitais para aplicação em sistemas com a planta nas configurações desacoplada e acopladas de fase mínima e não-mínima. Realizam-se ensaios via simulação, com o modelo não-linear da planta, bem como via experimentos práticos, com o sistema real, para validar os controladores projetados. Com isso, obtêm-se resultados positivos com ambas as técnicas estudadas, com exceção ao caso acoplado de fase não-mínima, para o qual uma das técnicas não foi capaz de controlar a atitude da planta. / This dissertation presents the mathematical modeling of a control moment gyroscope (CMG) and applies digital control techniques for its attitude control. CMGs are important actuators for bodies attitude control in the space, such as satellites and space vehicles. The system modeling is developed with the Lagrangian mechanics and a digital proportionalintegral- derivative (PID) and a digital linear-quadratic regulator (LQR) control techniques are studied and applied to a decoupled system and coupled systems with minimum and non-minimum phases. Simulation tests with the non-linear plant model and experimental tests with the real plant are done in order to validate the control techniques. Then, positive results are reached with both studied techniques, excepting in the coupled system with non-minimum phase, in which one technique was not able to control it.

Made available in DSpace on 2016-12-23T14:07:26Z (GMT). No. of bitstreams: 1 Alexandre Secchin de Melo - parte 1.pdf: 1376671 bytes, checksum: 7504ac977143d6adfd9ac0c7c2bc0d2f (MD5) Previous issue date: 2010-06-30 / Este trabalho visa a implementação de um objeto voador não-tripulado, em formato miniatura, com quatro rotores como plataforma de desenvolvimento, como parte de uma pesquisa mais abrangente. O objetivo final, ainda por ser alcançado, é chegar a um veículo voador miniatura com o máximo grau de autonomia de decisões baseadas no sensoreamento a bordo e poder computacional embarcados, estratégia do controle inteligente, e tarefa a cumprir. Tal implemetação, até o momento, consiste em uma máquina eletro-mecânica de baixo custo, cuja parte eletrônica a bordo, um microcontrolador de 8 bits, acelerômetros e giroscópios do tipo MEMS, permite a implementação de um controlador de voo genérico para automatizar a sua estabilização em torno dos eixos X, Y e Z. Tem vasta gama de aplicações como: inspeções aéreas em diversos ambientes, como linhas de transmissão elétrica, detecção de foragidos da polícia, monitoramento de plantações e rebanhos, bem como tomadas de filmagens para as indústrias cinematográfica e imobiliária. Testes realizados com o protótipo até agora sugerem a implementação bem sucedida de um controlador de estabilização de voo / This work aims the implementation of a unmanned, four-rotor miniature flying machine as a development platform, part of a long term research. The final goal, yet to be achieved, is the realization of a flying object with maximum practicable degree of autonomous decisions based on the on-board sensory and computational power, control strategy and assigned task. Up to this work, it consists of a low cost electro-mechanical hardware, whose electronic part allows the implementation of an 8-bit microcontroller-based, MEMS accelerometer and gyroscopes, allows the implementation of a generic fly control for attitude stabilization. The broad spectrum of applications includes: the inspection of various kinds of environments such as electric power transmission lines, police surveillance of woods and hard-to-reach places with sky sight for fugitive detection, crops and herd monitoring, as well as film takings for the cinematographic and real estate industries. Tests undertaking so far with the prototype suggest the successful implementation of a fly attitude controller

The goal of this project was to make a shirt with three embedded IMU sensors (Inertial Measurement Unit) that can measure a person’s movements throughout an entire workday. This can provide information about a person’s daily routine movements and aid in finding activities which can lead to work-related injuries in order to prevent them. The objective was hence to construct a sensor fusion framework that could retrieve the measurements from these three sensors and to create an estimate of the human body orientation and to estimate the angular movements of the arms. This was done using an extended Kalman filter which uses the accelerometer and magnetometer values to retrieve the direction of gravity and north respectively, thus providing a coordinate system that can be trusted in the long term. Since this method is sensitive to quick movements and magnetic disturbance, gyroscope measurements were used to help pick up quick movements. The gyroscope measurements need to be integrated in order to get the angle, which means that we get accumulated errors. This problem is reduced by the fact that we retrieve a correct long-term reference without accumulated errors from the accelerometer and magnetometer measurements. The Kalman filter estimates three quaternions describing the orientation of the upper body and the two arms. These quaternions were then translated into Euler angles in order to get a meaningful description of the orientations. The measurements were stored on a memory card or broadcast on both the local net and the Internet. These data were either used offline in Matlab or shown in real-time in the program Unity 3D. In the latter case the user could see that a movement gives rise to a corresponding movement on a skeleton model on the screen.

Today's state-of-the-art rate vibratory gyroscopes use a large proof mass that vibrates at a low resonance frequency (3-30 kHz), a condition that creates a performance tradeoff in which the gyroscope can either offer large bandwidth or high resolution, but not both. This tradeoff led to the development of the capacitive bulk acoustic wave (BAW) silicon disk gyroscope, a new class of micromachined rate vibratory gyroscopes operating in the frequency range of 1-10MHz with high device bandwidth and shock/vibration tolerance. By scaling the frequency, BAW gyroscopes can provide low mechanical noise without sacrificing the high bandwidth performance needed for most commercial applications. The drive loop of the BAW gyroscope can also be exploited as a timing device that can be integrated in existing commercial systems to provide competitive clock performance to the state-of-the-art using less area and power. This dissertation discusses the design and implementation of a CMOS ASIC architecture that interfaces with a high-Q, wide-bandwidth BAW gyroscope and the challenges associated with optimizing the noise performance to achieve navigation-grade levels of sensitivity as the frequency is scaled into the MHz regime. Mathematical models are derived to describe the operation of the sensor and are used to generate equivalent electrical circuit models of the gyroscope. A design strategy is then outlined for the ASIC to optimize the drive loop and sense channel for power and noise, and steps toward reducing this noise as the system is pushed to navigation-grade performance are presented that maintain optimum system power consumption. After analyzing the BAW gyroscope and identifying a strategy for developing the drive and sense interface circuitry, a complete fully-differential ASIC is designed in 0.18μm CMOS to interface with a bulk acoustic wave (BAW) disk gyroscope. As an oscillator, the gyroscope provides an uncompensated clock signal at ~9.64 MHz with a temperature sensitivity of -27 ppm/°C and phase noise of -104 dBc at 1 kHz from carrier. When the complete ASIC is interfaced with the gyroscope, the sensor shows a measured rate sensitivity of 1.15 mV/o/s with an open-loop bandwidth of 280 Hz and a bias instability of 0.095 o/s, suitable for the rate-grade performance commonly required for commercial and consumer electronics applications. The system is recorded to have a total power of 1.6 mW and a total area of 0.64 mm2. Following the design of the interface ASIC, this dissertation investigates in further detail the requirements for designing and optimizing charge pumps for capacitive MEMS devices. Basic charge pump design is outlined, followed by an overview of techniques that can be used to generate larger polarization voltages from the ASIC. Lastly, an alternate measurement technique for measuring the rotation rate of the gyroscope is discussed. This technique is based on the phase-shift modulation of the gyroscope output signal when the device is driven with two orthogonal signal inputs and can be easily modified to provide either linear scale factor measurement or a linear calibration curve that can be used to track and adjust the variation of the sensor scale factor over time.

For synthetic aperture imaging, position and orientation deviation is of great concern. Unknown motions of a Synthetic Aperture Sonar (SAS) can blur the reconstructed images and degrade image quality considerably. Considering the high sensitivity of synthetic aperture imaging technique to sonar deviation, this research aims at providing a thorough navigation solution for a free-towed synthetic aperture sonar (SAS) comprising aspects from the design and construction of the navigation card through to data postprocessing to produce position, velocity, and attitude information of the sonar. The sensor configuration of the designed navigation card is low-cost Micro-Electro-Mechanical-Systems (MEMS) Magnetic, Angular Rate, and Gravity (MARG) sensors including three angular rate gyroscopes, three dual-axial accelerometers, and a triaxial magnetic hybrid. These MARG sensors are mounted orthogonally on a standard 180mm Eurocard PCB to monitor the motions of the sonar in six degrees of freedom. Sensor calibration algorithms are presented for each individual sensor according to its characteristics to precisely determine sensor parameters. The nonlinear least square method and two-step estimator are particularly used for the calibration of accelerometers and magnetometers. A quaternion-based extended Kalman filter is developed based on a total state space model to fuse the calibrated navigation data. In the model, the frame transformations are described using quaternions instead of other attitude representations. The simulations and experimental results are demonstrated in this thesis to verify the capability of the sensor fusion strategy.

A new type of cantilever beam gyroscope is introduced, modeled, and analyzed. The main structure includes a cantilever beam and a rigid body attached to the free end of the beam. The model accounts for the eccentricity, that is the offset of the center of mass of the rigid body relative to the beam's free end. The first and second moments of mass and the rotary inertia appear in the equations of motion and boundary conditions. The common mechanism of electrostatic actuation of microgyroscopes is used with the difference of computing the force at the center of mass resulting in the electrostatic force and moment in the boundary conditions. By using the extended Hamilton's principle, the method of assumed modes, and Lagrange's differential equations, the equations of motion, boundary conditions, and the discretized model are developed. The generalized model simplifies to other beam gyroscope models by setting the required parameters to zero. Considering the DC and AC components of the actuating and sensing methods, the response is resolved into the static and dynamic components. The static configuration is studied for an increasing DC voltage. For the uncoupled system of equations, the explicit equation relating the DC load and the static configuration is computed and solved for the static configuration of the beam-rigid body in each direction. Including the rotation rate, the stationary analysis is performed, the stationary pull-in voltage is identified, and it is shown that the angular rotation rate does not affect the static configuration. The modal frequencies of the beam-rigid body gyroscope are studied and the instability region due to the rotation rate is computed. It is shown that the gyroscope can operate in the frequency modulation mode and the amplitude modulation mode. To operate the beam-rigid body gyroscope in the frequency modulation mode, the closed-form relation of the observed modal frequency split and the input rotation rate is computed. The calibration curves are generated for a variety of DC loads. It is shown that the scale factor improves by matching the zero rotation rate natural frequencies. The method of multiple scales is used to study the reduced-order nonlinear dynamics of the oscillations around the static equilibrium. The modulation equations, the ``slow'' system, are derived and solved for the steady-state solutions. The computational shooting method is employed to evaluate the results of the perturbation method. The frequency response and force response plots are generated. For combinations of parameters resulting in a single-valued response, the two methods are in excellent agreement. The synchronization of the response occurs in the sense direction for initially mismatched natural frequencies. The global stability of the system is studied by drawing phase-plane diagrams and long-time integration of response trajectories. The separatrices are computed, the jump phenomena is numerically shown, and the dynamic pull-in of the response is demonstrated. The fold bifurcation points are identified and it is shown that the response jumps to the higher/lower branch beyond the bifurcation points in forward/backward sweep of the amplitude and the excitation frequency of AC voltage. The mechanical-thermal (thermomechanical) noise effect on the sense response is characterized by using a linear approximation of the system and the nonlinear "slow" system obtained by using the method of multiple scales. To perform linear analysis, the negligible effect of Coriolis force on the drive amplitude is discarded. The second-order drive resonator is solved for the drive amplitude and phase. Finding the sense response due to the thermal noise force and the Coriolis force and equating them computes the mechanical-thermal noise equivalent rotation rate in terms of system parameters and mode shapes. The noise force is included in the third-order equation of the perturbation and equation to account for that in the reduced-order nonlinear response. The numerical results of linear and reduced-order nonlinear thermal noise analyses agree. It is shown that higher quality factor, higher AC voltage, and operating at lower DC points result in better resolution of the microsensor.