A first approach to a Simultaneous Localisation and Mapping (SLAM) solution implementing the Extended Kalman Filter for visual odometry data.
This project was a done as a _nal grade project to obtain the degree in communications electronics engineering from TECNUN. It was done at Vicomtech under the supervision of Leonardo de Maeztu , Marcos Nieto and Ainhoa Cortes. This project has consisted on the study and a _st approach implementat...
| Main Author: | |
|---|---|
| Format: | info:eu-repo/semantics/bachelorThesis |
| Language: | eng |
| Published: |
2015
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10171/37574 |
| Summary: | This project was a done as a _nal grade project to obtain the degree in communications
electronics engineering from TECNUN. It was done at Vicomtech
under the supervision of Leonardo de Maeztu , Marcos Nieto and Ainhoa Cortes.
This project has consisted on the study and a _st approach implementation of
a simultaneous localisation and mapping algorithm. The algorithm of choice
was the EKF SLAM(Extended Kalman Filter Simultaneous Localisation and
Mapping).To properly understand the concepts behind SLAM and its implementation
various papers and project reports have been read, but the theoretical
implementation and the practical implementation have been mostly based
on the report by Jose-Luis Blanco,\Derivation and Implementation of a Full
6D EKF-based Solution to Bearing-Range SLAM"[9]. Also for the implementation
, the software C++ has been used with the help of the OpenCV library
for the handling and processing of images and matrices. Matlab was also used
when complicated math operations needed to be done. The implementations
was made with the help of previous code provided by Vicomtech [1] . The code
provided contained a successful implementation of a visual odometry problem
and it included most of the image processing needed for the next steps of this
project. |
|---|