Self-localization of mobile robot in unknown environment
- Resource Type
- Conference
- Authors
- Prozorov, A.; Tyukin, A.; Lebedev, I.; Priorov, A.
- Source
- 2015 17th Conference of Open Innovations Association (FRUCT) Open Innovations Association (FRUCT), 2015 17th Conference of. :173-178 Apr, 2015
- Subject
- Communication, Networking and Broadcast Technologies
Components, Circuits, Devices and Systems
Computing and Processing
Geoscience
Signal Processing and Analysis
Cameras
Simultaneous localization and mapping
Robot kinematics
Particle filters
Robot vision systems
- Language
- ISSN
- 2305-7254
In this paper we propose a method for solving the SLAM problem for mobile robot when moving in an unknown environment. Our method takes computational advantages of the FastSLAM algorithm. To estimate the position of the robot, we use a particle filter. The weights for the set of particles that characterize the expected position of the robot, are determined by the condition number of the plane homography matrix. It can be considered as the projective mapping of points of the scene on the two-dimensional surface of camera sensor. A set of unscented Kalman filters is used to estimate the positions of detected landmarks which are forming the map of the observed environment. Methods for detecting and description of landmarks were not considered in this paper, as it is beyond the scope of this work.