Home
About Us
People
Visitors
Groups
Projects
Publications
Software
Courses
Laboratory
Seminars
Students
Matlab
Alumni
Links
 
[Advanced Control Systems Group] [Autonomous Mobile Robotics Group]





Mobile Robots Localization and Map Building


Simultaneous Localization And Map building


Simultaneous Localization and Map Building (SLAM) problem occurs when a mobile robot autonomously navigate in an unknown environment. The robot starts in an unknown location in the environment and builds its map and simultaneously relocates itself by using observations of its on-board sensors only. A solution to SLAM problem makes a mobile robot really autonomous because it eliminates the need of knowing robot's absolute location or a priori map of the environment. The solution to SLAM problem consists of the following stages: estimation of nonlinear and stochastic systems, mathematical modeling of sensors and environment, processing sensor information and pattern recognition. In our first solution to SLAM problem, we use exact mathematical approach, namely direct stochastic approach, in terms of Symmetries and Perturbation model (SPmodel) and fusion of laser rangefinder and monocular vision data using Extended Kalman Filter.


Demonstration of the EKF based SPmodel SLAM solution with LRF is given in below movie, which shows how the consecutive scans of the environment are integrated into a global map and how the mobile robot updates its pose and global map when it establishes correspondences between local map features and the global map features.
Launch in external player

 

Home
About Us
People
Visitors
Groups
Projects
Publications
Software
Courses
Laboratory
Seminars
Students
Matlab
Alumni
Links