Control of ardupilot quadrotor : Part III
The basic principle for estimating the 3-dimensional parameters from a sequence of images comprises of two main problems. The first problem is the estimation of the moving object’s parameter. The second problem is generating the range information assuming the motion parameters are known. In this...
Saved in:
Main Author: | |
---|---|
Other Authors: | |
Format: | Final Year Project |
Language: | English |
Published: |
2013
|
Subjects: | |
Online Access: | http://hdl.handle.net/10356/53346 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
Summary: | The basic principle for estimating the 3-dimensional parameters from a sequence of images comprises of two main problems. The first problem is the estimation of the moving object’s parameter. The second problem is generating the range information assuming the motion parameters are known.
In this project, the primary focus is the perspective non-linear observer to solve the range identification problem. TLD (Tracking-Learning-Detection) and Identifier Based Observer (IBO) were chosen to solve the two problems respectively.
TLD is an effective method to generate the image coordinates of the tracked object. IBO performance was responsive for simulations. However, when IBO was implemented for experiment, the performance differs. This might be due to the object moving speed and the limitations of the indoor environment. |
---|