The Research Of RFID In Ball Location Of The Intelligent Ball-picking Robot

With the continuously upgrade of life level and science and technology level, intelligent robots are gradually entering people's lives. Today, more and more people pay attention to the health, so physical exercise already becomes a part of daily life. Table tennis and tennis etc sports are loved by people, because they are convenient and quick. But consequential peoples have to pick up the balls of scattered in the stadium. In order to solve this problem, researchers have developed an intelligent ball picking robot replace the manual to picking up the ball. But there has a great development space on sphere orientation field and RFID indoor positioning technology is the hotspot of research.At present, indoor positioning technology of based on RFID mainly uses LANDMARC algorithm and determine the physical location of to be positioning label by introducing an auxiliary reference label, it has ideal of fixed position accuracy in the indoor environment. But the LANDMARC algorithm is insufficient, the positioning accuracy relevant to the numbers of neighbor reference tags and the density of reference tags. So this paper based on the experiment analysis conducted to research the sphere positioning technology based on LANDMARC algorithm. Include the following two aspects:(1)Improve the algorithm about the positioning accuracy of the LANDMARC algorithm relevant to the numbers of neighbor reference tags by introducing the weighted idea in the neighbor reference tags. The traditional algorithm is limited to small range positioning environment for selecting the 3-5 reference tags, from the reader far adjacent reference tag interference selection of neighbor reference tag in the context of a larger location, so need to select more reference tags to the auxiliary positioning. But the positioning error will be increased with increasing the number of the selected reference tags. The improved LANDMARC algorithm optimized the weight distribution of calculation process and reduced the system error and improved the positioning accuracy. Simulation results show that the improved algorithm has high precision in many cases selected reference tags.(2)Because the algorithm of LANDMARC positioning accuracy related to the density of tag deployment, so introduction the virtual reference tags to improve positioning accuracy. The virtual reference tags and real reference tags have the same characteristics that the receiving field strength value can calculation according to the path loss model. Therefore, carries on two times positioning by combination the positioning algorithm of virtual reference tags and the improved LANDMARC localization algorithm. Simulation results show that localization algorithm based on virtual reference tags improved positioning precision and has obvious effect for a single to be positioning label. Ball positioning precision level has increased from dm to cm.