Loading...
| Friend's email | |
| Your name | |
| Your email | |
| enter code | |
This page was sent successfuly
- Type of Document: M.Sc. Thesis
- Language: Farsi
- Document No: 44318 (08)
- University: Sharif University of Technology
- Department: Mechanical Engineering
- Advisor(s): Alasti, Aria
- Abstract:
- Algorithms for online identification of parameters of a satellite’s moment of inertia tensor based on two methods (recursive least squares and Kalman Filter) is presented and compared with each other. It is assumed that the satellite actuators are three orthogonal reaction wheels. The only available sensor is a 3-axis rate gyro, which measures the angular velocity of satellite in body coordinate system. For validation of algorithms on earth, a similar identifier was designed for a satellite simulator. Due to difference in Dynamics of satellite and satellite simulator, capability of identifying center of gravity was added to algorithms. In order to account for friction in air bearing of satellite simulator, some friction compensation strategies was used. Due to existence of noise in sensor, the regressor matrix used in least squares method, changed stochastically. Therefore, the classic least squares method would not converge and was not useful. To resolve this problem, a modified least squares method with robust scheme was provided and its stability was proved using Lyapunov stability theory. Simulation results showed that extended Kalman filter-based algorithm could identify inertia parameters with less than 5 percent error in presence of 5 percent noise, while least squares-based algorithm could do the same in presence of 0.1 percent noise. Meanwhile, sensitivity of identifiers to noise and bias in disturbances was studied and Kalman filter method showed better performance
- Keywords:
- Low Earth Orbit (LEO)Satellite ; Recursive Least Square Estimation ; Extended Kalman Filter ; Parameters Identification ; Inertial Tensor Moment ; Satellite Simulator
Digital Object List
-
محتواي کتاب
- view
Bookmark