Semantic-understand-based landmark navigation method of robots

In this paper, we propose a semantic-understand-based, mapless navigation method for robots, which directly using the human navigation system - landmarks. This method locates and tracks landmarks via a color-based region segmentation algorithm, and when within proper distance, extracts texts from regions of landmark and inputs them into the OCR engine Tesser-act [1] for recognition. Simultaneously, a projection analysis of arrows on the landmark is conducted, identifying the semanteme of arrows. Finally, by combining the semanteme of arrows and texts which extract from landmark, the robot can find the routes to the destinations automatically. Experiments on P3AT robot show that our method can used to accomplish mapless navigation task perfectly in indoors environment. Being able to navigate with landmarks in real life directly without generating maps or resetting new navigation sign for robot specially, our method can be applied in the field of service robots rapidly, which can enhance their adaptability and viability significantly.