Curso Dr. Miroslav Kulich

Mobile Robotics II: Simultaneous localization and mapping

 

Summary

The aim of Mobile Robotics II is to introduce one of the hottest topics in mobile robotics – simultaneous localization and mapping (SLAM). SLAM is a technique to build a model (map) within an unknown environment by a mobile robot while at the same time keeping track of robot’s position. The course is divided into two parts. In the theoretical part, the students will learn fundamental methods of si- multaneous localization and mapping. In the practical part, the students will imple- ment the learned methods and test them in a simulated environment as well as on real robots. The practical experiments will be done with a e-learning system SyRoTek (sy- rotek.felk.cvut.cz) developed at Czech Technical University and introduced during last ECI course. The course is offered based on response and requests of participants of Introduction to mobile robotics course. The knowledge of topics learned during last year are welcome but not necessary.

 

Schedule

  1. Theory - probabilistic methods in robotics, processing of uncertain data, Bayes filter, Kalman filter, Extended Kalman filter, particle filter
  2. Theory - SLAM problem definition, Rao-Blackwellized filter, GraphSLAM, FastSLAM, EKF-SLAM, correspondence problem, loop closing problem (gathering and processing sensor data)
  3. Labs - introduction to ROS (Robot Operating System) and SyRoTek system, task defini- tion, solving simple example in SyRoTek system
  4. Labs - individual solving of the task under supervision. The students will be split into groups of three or four and each group will choose one task from the pool of task pre- pared by the teacher. The tasks will be focused on a particular sub-problems of SLAM (correspondence problem, sensor or motion model) or particular SLAM algorithms.
  5. Labs - individual solving of the task (in groups) under supervision. Continuation of the work from the previous day. It is expected that students will continue their work also few days after the lab to make a working solution.

 

Bibliography

• Thrun, S.: Probabilistic algorithms in robotics. AI Magazine, 21(4):93-109, 2000. 

• Slides of SLAM Summer School 2009: http://www.acfr.usyd.edu.au/education/summerschool.shtml 

• web pages of OpenSLAM.org: http://www.openslam.org/ 

• web pages of Robot Operating system: http://ros.org 

• web pages of SyRoTek system: syrotek.felk.cvut.cz