Sharif Digital Repository

The goal of active learning is to choose the training instances intelligently, in order to make the cost of the learning process as low as possible. In proactive learning, which is an extension of active learning, several noisy labelers provide labels for each chosen sample. The proactive problem consists of three different subproblems: estimating true labels and finding their correctness probability (crowdsourcing), learning a model for finding labels of new instances (classification), and selecting samples for empowering the classifier (active learning). Since estimated labels through crowdsourcing are not certain, regular classification and active learning methods should be extended to be... 

One of the approaches used in learning theory is using information theoretic tools. The general idea of this approach is that if we show the algorithm did not memorize the dataset, we could guarantee generalization. Noisy channel model is one of the important concepts in this approach. A noisy channel is a lossy process which maps the data to a compressed format.There are two ways to use noisy channel model in literature: input compression and model compression. One of the main results of this thesis is to show that the input compression methods can not explain the generalization of algorithms (despite previous belief). On the contrast by fixing some of the problems faced in the model... 

Accumulative Roll Bonding (ARB) process has been developed during last decade to produce the Ultra Fine Grained (UFG) structures with high strength. In the present study, a Ti-IF steel was deformed up to equivalent strain of 4.8 at 500 ˚C using ARB process. For comparing the ARB with conventional forming processes, some samples were deformed with similar equivalent strain to ARB conditions at 500 ˚C and 300 ˚C as warm rolling. Moreover, the thermo-mechanical warm rolling was simulated by warm torsion. The ARB and warm rolled samples with largest strains were annealed at 750 ˚C. The microstructure and texture were studied using Electron Back Scattered Diffraction (EBSD) method. The... 

Aerodynamics loads and aeroheating during reentry, are one of the challenging issues in designing One of the main challenges in the design and application of reusable spacecrafts is the aerodynamic and thermal loads acting on their structures during reentry into the atmosphere. For this reason, all the concepts/prototypes presented in the 20th century were either not fully reusable, or were partially canceled, or did not have economic justification. In this report, the aerodynamic loads on the first stage of a launch vehicle during the launch and return phases have been investigated. Due to the fact that the main part of the first stage of the vehicle includes fuel and oxidizer tanks, the... 

The collision of moderately large asteroids and comets also referred to as Near-Earth Objects (NEO's) with earth would have catastrophic consequences. Such events have occurred in the past and may occur again in the future. However, for the first time in known history, humanity may have the technology required to counter this threat. Methods studied for mitigation of this hazard are based on deflecting asteroid's trajectory and are normally divided into two basic categories: high energy impulsive methods, and long-duration low-thrust methods. Two parts of mission to be optimized are the trajectory of spacecraft to reach the target and the deflection strategy which could be analyzed by the... 

The main purpose of the present thesis is to establish a decentralized controller for some car–like multi robots to follow and hunt moving targets. Robots are very similar to actual cars considering geometric dimensions, mass and moment of inertia and so on; and outputs of the controller are steering-wheel and driving-wheel torques appropriately. Dynamics of the moving target is so that it escapes from the robots. Robots are equipped with antenna for getting wireless sensory signals, including range and bearing sensors. Kalman filter is used for estimation the target relative position and speed and robots state variables. The controller is designed for doing the above mentioned group... 

This research proposes a novel hybrid method for stock price index forecasting relying on technical analysis, the fundamental analysis of capital market, neural networks technology, and fuzzy logic. The new method shows a good performance comparing complicated and time-consuming forecasting methods. Several factors influence stock price; for example, an important one is the previous trend of stock price. Fortunately, by developing technical analysis and introducing the various indices of this method the maximum use of historical data is made in order to forecast future prices. Another kind of factors is macro-economic variables that their influence on the long-term trends of the stock... 

In this thesis control of wheeled mobile manipulators for cooperative object transportation is investigated. Full dynamic model of a mobile manipulator with a three wheeled mobile base and a three DOF manipulator is derived using Gibbs-Appell method. Input-output linearization technique is used to control individual robots. Transporting the object is divided into two steps, first robots use a decentralized behavior-based method to approach and surround the object. Then virtual structure method is used to control the robots to transport the object cooperatively. Simulation results with three robots handling an object are presented to show the effectiveness of proposed method  

In this work dynamic behavior of a two body TSS with main satellite as rigid body is studied and effects of tether elasticity and tension on the satellite dynamic are indicated in station keeping phase and circular orbit. Also a simplified modeling is introduced which could be used in nonlinear controller design. Lagrange and extended Lagrange method is used to derive equations of motion. It is shown that longitudinal vibration of tether has small effect on satellite attitude and angular velocity and because of high frequency it can be decoupled from satellite dynamic. In the simplified model, orbital motion is decoupled from equations of TSS and main satellite dynamic. Effect of decoupling...