Improvement on the Image Processing for an Autonomous Mobile Robot with an Intelligent Control System

A robust and reliable path recognition system is one necessary component for the autonomous navigation of a mobile robot to help determining its current position in its navigation map. This paper describes a computer visual path-recognition system using on-board video camera as vision-based driving assistance for an autonomous navigation mobile robot. The common problem for a visual system is that its reliability was often influenced by different lighting conditions. Here, two different image processing methods for the path detection were developed to reduce the effect of the luminance: one is based on the RGB color model and features of the path, another is based on the HSV color model in the absence of luminance.