Bridging the Gap Between Feature- and Grid-based SLAM

This experiment illustrates a SLAM approach which uses a combination of occupancy grid maps and feature-based models (poles in this example). In every timestep the robot uses the representation which explains the data best. This is a project of Kai M. Wurm, University of Freiburg. Details of the approach have been published in the Journal of Robotics & Autonomous Systems (2009) and at the European Conference on Mobile Robotics 2007: http://www.informatik.uni-freiburg.de/~wurm/papers/wurm07ecmr.pdf

Category:

Science & Technology