Color Landmark Design for Mobile Robot Localization

Abstract

For the fast and accurate self-localization of mobile robots, artificial landmarks can be used very efficiently in the complex workspace. In this paper, we designed a color artificial landmark with symmetric rectangles and seven-part numbers, which showed cross- ratio invariance under perspective transformation and proposed a fast landmark detection-recognition algorithm based on the proposed landmark model. Detection- recognition of the model was accomplished by color information and cross-ratio invariance. We also proposed a self-localization algorithm based on the landmark model. We demonstrated the feasibility of the proposed technique through practical experiments in cluttered indoor environments and simulation experiments. Experimental results show that proposed landmark is enough to be used in cluttered environment and landmark detection-recognition in cluttered scene is real-time robustly under various viewing angles, self-localization is accurate in the presence of additive random noise.