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: 49502 (19)
- University: Sharif University of Technology
- Department: Computer Engineering
- Advisor(s): Manzuri, Mohammad Taghi; Jafari, Mahdi
- Abstract:
- Today with increasing advancement of robotic technology, we witness the increment of robots role in humans’ life. One of the issues always present for robots is routing which may require locating their position. Most of these robots make use of sensors like gyroscope, accelerometer and GPS to estimate their position. But each of these sensors has their own short comings such as drift or being unavailable in certain environments. Fusion of inertial and vision sensor’s data can be used to reduce the impact of these errors. Considering the time consumption of image processing and accumulating error of inertial sensor, this combination can be very useful. This approach is also used in smartphones, cameras, etc. the challenge here is that when inertial sensor starts operating, it produces a random number as bias that causes inconsistency between sensors. The goal of this project is to achieve an appropriate estimation of rotation and increase the accuracy by using an optimized filter and calculating angle with combination of vision and gyroscope. For this purpose, we calculated camera orientation differentiation with error of 0.02 degree and 1.66 times improvement in calculation time by implementing a Particle filter
- Keywords:
- Optical Filters ; Sensor Fusion ; Position Estimation ; Optimal Filtering ; Sensor Error
Digital Object List-
محتواي کتاب - view
Bookmark