Evolutionary Parameter Optimization for Visual Obstacle Detection
Thomas Bergener, Carsten Bruckhoff, Christian Igel · 2000
In this paper we employ an Evolutionary Algorithm (EA) to improve the parameters of a visual obstacle detection method called Inverse Perspective Mapping. We show that the EA leads to a better parameter setting than the one found by an expert. The obstacle detection method is successfully implemented on our autonomous mobile robot ARNOLD to navigate in an unknown and dynamically changing environment in a fast and reliable manner.