BibTeX
@MISC{Röfer95controllinga,
author = {Thomas Röfer},
title = {Controlling a Robot With Image-based Homing},
year = {1995}
}
Bookmark
 |
 |
 |
 |
 |
 |
OpenURL
Abstract
. This paper presents a biologically inspired method for the navigation of autonomous mobile systems. The method calculates the way from a current position to a target position using one-dimensional 360 images, taken at these positions. The correlations between the two images are generated by using a modified version of Kohonen's self-organizing feature map. The direction to the target position is calculated from the weight vectors of the trained network. Experiments with a small robot prove the method to be stable and exact. 1 Introduction The navigation of autonomous mobile robots is still an unsolved problem. Robots operate in a three-dimensional, changing world. In this environment they have to navigate in real time. For this reason there has to be a suitable technique for navigation. Methods that are based on three-dimensional models are very complex: it takes a lot of time to match the sensor data with three-dimensional objects stored in large databases. Therefore it see...
Citations
No citations identified.
