Sharif Digital Repository

This research focuses on the robust mobile robot localization exploiting motion information acquired from an optical mouse operating based on optical flow technology. Most techniques of visual motion measurement are based on the well research discipline called “optical flow”. Theoretically, optical flow as a method of localization can be highly accurate, but it is sensitive to the noise and surface texture/optical characteristics and distance variations between the CCD detector and surface. As one could not achieve acceptable results in practical situations, to handle these problems, we propose to attach an acceleration – gyro (INS) sensor on the CCD detector (optical mouse) to improve the... 

There are challenges in controlling and guidance of vehicles which makes it necessary to have true information about the position, velocity, and orientation of the system. This issue is severe to some extent when a UAV is the object to be controlled. The navigation system is responsible for supplying the mentioned information which is usually accompanied with some errors. Therefore, threre are a wide range of algorithms developed for omitting these errors leading to a well-calibrated navigation system. One of the precise and reliable calibration method is based on the data fusion of intertial navigation sensors and magnetometer measurements. As the obtained data via magnetometers are usually... 

In this thesis, in order to improve the performance of the mobile robot navigation, a Fuzzy approach is used for making a safe path in the cluttered environment with hurdles in the work space of the robot. To this end, based on data collected from instantaneous location of the robot and location of the robot, heuristic rules are extracted. Also, in order to obtain optimal data fusion of the sensors, Kalman filter is used to localize the robot. In this regards, by using the kinematics of the robot and supposing the white noise in the process and measurements, the position and orientation of the mobile robot are estimated in a real-time and adaptive manner  

In this project a framework for Model Based Fault Diagnosis (MBFD) in a Gas Transmission Network (GTN) was developed. A black-box linear state-space model was used to capture dynamic behavior of an industrial benchmark for GTN. A full fledge commercial network simulator were used to obtain the required data for process identification. In order to check the robustness of the model zero mean white noise was imposed on various output obtained by the commercial simulator. Kalman filter was used to estimate the states of the GTN. These estimated states along with the measured output (all obtained from the commercial simulator) are almost similar to their corresponding estimated signal. This shows... 

Simulation of cardiovascular system functionality during various physiological conditions is essential at different diagnostic and clinical levels. A first step in studying the roots of cardiovascular diseases and abnormal activity is to study a practical yet complete model of the cardiovascular system. In this thesis we introduced a new approach for defining the distributed model of the cardiovascular system. Initially, we chose an appropriate subsystem, namely Arch of Aorta, and proposed a distributed model for it. The elements of the proposed model were nonlinear RLC elements that simulate resistance, blood viscosity and vessel elasticity respectively. To minimize the system complexity,... 

Control and guidance of a vehicle is vital for it to do its mission correctly. Guidance system and decisional algorithms used in it need exact information of instantaneous position, velocity, and attitude of the guided vehicle to determine and issue proper guiding commands.
Estimation algorithms and navigation sensors are two necessary tools for navigation. In the first section of present study, the satellite equations of motion are derived and a 6-DOF simulation is done for it. In the second section, measurements of star tracker will be simulated using results of the first section. The third section appertains to development of navigation equations, attitude determination using UKF... 

Modeling and Controlling of a satellites fuel slosh is considered a vital problem in the field of satellite attitude control. In this paper modeling and control of the attitude of a satellite carrying a partially filled spherical fuel tank, which uses the diaphragm as a propellant management device (PMD), has been studied. The fuel slosh dynamics is modeled using a spherical pendulum with two degrees of freedom and a rigid arm, which represents the first mode of sloshing. The effects of the diaphragm that surround the fuel surface, are considered as torsional springs and dampers attached to the satellite body. The combined dynamics equations of the satellite rigid motion and the fuel slosh... 

In this Thesis, attitude and orbit of a satellite is estimated using a star sensor and an earth sensor. An algorithm has been developed to estimate satellite attitude and orbit that can be implement on a satellite as strap-down hardware. This system has an uncomplicated implementation compared with platform based systems. That mentioned algorithm estimates satellite orbit by using identified star vectors via star sensor and identified earth vector via earth sensor simultaneously. It also knows the stars vectors according to Star Catalog. Attitude estimation is also executed using both star and earth sensor. It has to be said that an Unscented Kalman Filter synthesize both sensors output data... 

Nonlinear nature of polymerization reactors has always attracted great attentions in various scientific areas such as modeling, simulation, dynamical behavior investigation and control engineering as well. Free radical polymerization in continuous reactors, whether in terms of design and startup or in terms of economical concerns, has numerous fans in industrial aspects. In the following thesis, the dynamical behavior of continuous stirred tank reactor of Vinyl Acetate free radical polymerization is examined in an industrial scale. A control system with optimum performance is designed for this CSTR as well. According to the common behavior of this type of reactor, it may demonstrate... 

The conventional bearing-only tracking solutions are based on Kalman filter. Whenever the problem equations are nonlinear, Extended Kalman Filter (EKF) or more complicated methods are used necessarily. However, EKF suffers of being biased and unstable. Here a new recursive method based on a linear modeling of the problem is proposed, the method does not need an initial guess and outperforms the Kalman filter in the terms of accuracy, complexity and stability. Furthermore, input information needed for the proposed tracking filter is just the observer position, time and value of each angle measurements and the computation load is less than even a simple EKF. mathematical analysis, as well as... 

Most of the existing visual SLAM methods heavily rely on a static world assumption and easily fail in dynamic environments. One solution is to eliminate the influence of dynamic objects by introducing deep learning-based semantic information to SLAM systems. In this project, we propose a real-time semantic RGB-D SLAM (built upon RTAB-Map) system for dynamic environments that is capable of detecting moving objects and maintaining a static map for robust camera tracking. Furthermore, we augment the semantic segmentation process using an Extended Kalman filter module to detect temporarily static moving objects by adding centroids to each found dynamic object and calculating their velocity. We... 

Many advanced process plants and machines are extremely complex, and always depend on automatic control for satisfactory operation. In order to achieve and maintain system stability and assure satisfactory and safe operation, there is increasing demand for dynamic systems to continue acceptable operation following failures. Therefore failures detection has always been an important aspect of fault tolerant control design. In nuclear power plant, for example, tens of alarms can occur in a few second after a fault. Fault detection might be of utmost importance due to safety, political and other reasons. There are many methods to detect fault on the system. In this project approach based on... 

Hydrocarbon reservoir management and development as well as planning of enhanced oil recovery (EOR) processes are based on the reservoir dynamic model. Thus, successful implementation of EOR scenarios greatly depends on the quality of the dynamic model and accuracy of the associated parameters in order to correctly describe fluid flow through porous media. First, a dynamic model is constructed based on the prior knowledge. However, because of the various types of error during model building, the prior model is not so accurate and perfect. Accordingly, new observation data, such as production and 4D seismic data, are utilized to calibrate the prior model and characterize the reservoir under a... 

This paper puts forward a probabilistic approach for simultaneous identification of the dynamic stiffnesses of the soil-foundation and the foundation input motion of soil-structure systems from sparsely-measured structural responses. The soil-structure system in this approach is represented by a shear building resting on sway- and rocking-representative springs and dashpots. The proposed approach employs unknown-input extended Kalman filtering in time domain. This method is built upon Bayesian updating and accounts for two major sources of uncertainty in the problem: the process noise that account for the modeling errors and the measurement noise. In the proposed method, the state vector of... 

The aim of this article is ancestral control of a set of 3-wheel robots with non-holonomics dynamics for hunting and covering around moving target. Robots have mass and inertia and the wheels' mass have been considered in the dynamics of robots. The output of controller is wheel's torque and front wheel's steering torque. Moving target dynamics has been considered in a way that it is escaping from invader and the target has holonomics dynamic. To derive the equation of motion, Kane's dynamics procedure has been used. Robots are equipped to sensors for distance assessment, vision angle assessment and also signal receiver antenna. To estimate relative position and variables of other robots'... 

In this project, a processing algorithm is approached which is resistant against environmental and motion noises to find the pulse transmission time using two photoplethysmography signals. For this purpose, a new processing framework is checked and generalized for photoplethysmography signals. Already, this processing framework had impressive results. The considered processing method is obtained by offering a dynamic model for the signal and using it in a Kalman filter structure. For the photoplethysmography signals, with modeling every beat of signal to a Gaussian three or four sum form and with adding self-returned equations for model parameters, a nonlinear signal model is obtained. Then,... 

Robust detection of human eye has attracted interest during the past decade. Certain research studies have focused on the capability of determining the location of the human eyes while few have been focused on its tracking. The main task in this study is not only a system with eye detection capability but also a robust and non-intrusive eye tracker system of the eye movement using the iris as a trace element within the image captured through the camera. Such utilization is particularly applicable in diagnosing of certain abnormalities such as Autism which result in distinct abnormal movement and behavior in the eyes. Proposed eye tracker algorithm was implemented using the Kalman filter. The... 

Hydrocarbon reservoirs development and management is based on their dynamic models. To encounter various types of error during model building, model parameters are adjusted to produce reservoir historical data by assimilation (history matching) of reservoir production or 4D seismic data. Among the existing sequential methods for automatic history matching, ensemble Kalman filter and its variants have displayed promising results. The innovations of this thesis for ensemble Kalman filter (EnKF) are presented into three major orients; these includes adaptive localization/regularization, characterization of original PUNQ test model and characterization of channelized reservoir.
To mitigate... 

Nanopositioning is the technology to control position of the objects with precision of the scale of nanometers. Nowdays, This technology emerged as one of the components of research and production in the micro and nanometer dimensions, helping researches and producers to carry out many state of the art projects. A nanopositioning device, consists of a support to path guidance, an actuator to apply force and motion, a driving circuit for communication with the actuator, a sensor for measuring displacement, a circuit for reading sensor’s data, a controlling algorithm to reach the desired precision, a controlling hardware to enforce controlling logic, a power supply and a graphical user... 

During the last years, a more volatile and dynamic financial environment has caused an increasing concern about the stability of banking systems. In this sense, it is widely agreed that credit risk is one of the variables that are more directly related to financial stability. One of the most important purposes of modeling credit risk, is estimating credit loss distribution and forecasting expected loss. In this research, we estimate and analysis credit loss by considering macroeconomic variables and latent factors. We express loans losses in terms of four stochastic components: default frequencies, the size of the loans portfolio, the exposures at default and the losses given default. ...