Visual simultaneous localisation and map-building supported by structured landmarks

Authors:
Robert Bączyk;Andrzej Kasiński
Affiliations:
Institute of Control and Information Engineering, Poznań University of Technology, Pl. Marii Skłodowskiej-Curie 5, 60-965 Poznań, Poland;Institute of Control and Information Engineering, Poznań University of Technology, Pl. Marii Skłodowskiej-Curie 5, 60-965 Poznań, Poland
Venue:
International Journal of Applied Mathematics and Computer Science
Year:
2010

Citing 10
Cited 0

Simultaneous Localization and Map-Building Using Active Vision

IEEE Transactions on Pattern Analysis and Machine Intelligence
Estimation with Applications to Tracking and Navigation

Estimation with Applications to Tracking and Navigation
Real-Time Simultaneous Localisation and Mapping with a Single Camera

ICCV '03 Proceedings of the Ninth IEEE International Conference on Computer Vision - Volume 2
Stanley: The robot that won the DARPA Grand Challenge: Research Articles

Journal of Robotic Systems - Special Issue on the DARPA Grand Challenge, Part 2
MonoSLAM: Real-Time Single Camera SLAM

IEEE Transactions on Pattern Analysis and Machine Intelligence
Edge landmarks in monocular SLAM

Image and Vision Computing
Real-time model-based SLAM using line segments

ISVC'06 Proceedings of the Second international conference on Advances in Visual Computing - Volume Part II
Guest Editorial Special Issue on Visual SLAM

IEEE Transactions on Robotics
Inverse Depth Parametrization for Monocular SLAM

IEEE Transactions on Robotics
Discovering Higher Level Structure in Visual SLAM

IEEE Transactions on Robotics

Quantified Score

Hi-index	0.00

Visualization

Abstract

Visual simultaneous localisation and map-building systems which take advantage of some landmarks other than point-wise environment features are not frequently reported. In the following paper the method of using the operational map of robot surrounding, which is complemented with visible structured passive landmarks, is described. These landmarks are used to improve self-localisation accuracy of the robot camera and to reduce the size of the Kalman-filter state-vector with respect to the vector size involving point-wise environment features only. Structured landmarks reduce the drift of the camera pose estimate and improve the reliability of the map which is built on-line. Results of simulation experiments are described, proving advantages of such an approach.