Position estimation method for agricultural field road

This paper introduces an artificial landmark position estimation method using omnidirectional vision for agricultural vehicles field road navigation. This system is built with an omnidirectional vision sensor, two landmarks, PC and image processor. We proposed the algorithm extracted landmark candidate in the image, and estimated the directional angle between landmarks and the distance between landmark and omnidirectional vision sensor. Based on the directional angle and the distance, the algorithm estimated the absolute position of the sensor relative to the landmark-based coordinate system on the ground. Experimental results show that the RMS distance error is about 15 cm on a 20 m distance experiment. We conclude that the proposed self-localization method is feasible and effective for agricultural vehicles field road navigation.