Extended Kalman Filter (EKF) Based Localization Algorithms for Mobile Robots Utilizing Vision and Odometry

In this paper, we describe a positioning method for a moving mobile robot in a known environment. The proposed technique incorporates location estimation by letting cameras to recognize the QR code from fixed markers. These fixed points will provide the reference points listed in the database. The s...

Full description

Saved in:
Bibliographic Details
Main Authors: Tran, Hoang T., Vo, Thanh C., Tran, Dong Lt, Nguyen, Quan Na, Ha, Duyen M., Pham, Quang N., Le, Thanh Q., Nguyen, Thang K., Do, Hai T., Nguyen, Minh T.
Format: Conference Proceeding
Language:English
Subjects:
Cameras
Data fusion
Extended Kalman filter
GPS
Laser range finder
Localization
Navigation
Omni-camera
QR codes
Robot localization
Robot vision systems
Sensor fusion
Sensors
Service robots
Sonar
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this paper, we describe a positioning method for a moving mobile robot in a known environment. The proposed technique incorporates location estimation by letting cameras to recognize the QR code from fixed markers. These fixed points will provide the reference points listed in the database. The system will calculate the angle to each landmark and then correct the robot's directions. The extended Kalman filter is deployed to correct the position and orientation of the robot from the error between the viewing angle and the estimate to each datum. The experimental results show that the approach improves and suffices in robot localization for navigation tasks. Results from experiments in real environments are presented including analysis. The results are significant and show promise.
ISSN:2158-8481
DOI:10.1109/MELECON53508.2022.9843066