A collision-aware mobile tag reading algorithm for RFID-based vehicle localization

With the development of radio frequency identification (RFID) technology, RFID systems on roads (RSR) have recently been put forward to localize not only manned vehicles but also autonomous vehicles. In such an RFID-based localization infrastructure, passive RFID tags with road-related information are deployed on road surfaces, and any vehicle equipped with an RFID reader can interrogate the tags to obtain its locations. However, like traditional RFID systems, reader collisions could occur when many vehicles (readers) continuously move in and out of a tag's reading range, significantly degrading tag reading performance. Although many anti-collision algorithms have been devised to improve tag reading efficiency, none can be applied in highly mobile and dynamic road environments. In this paper, we propose an adaptive mobile tag reading algorithm, which can effectively reduce reader collisions and improve tag reading efficiency for reliable vehicle localization under various traffic scenarios. Extensive simulation and experiment have verified the feasibility and the effectiveness of our proposed scheme. Particularly, the field tests demonstrate that, with moderate vehicle density and sparsely deployed tags, our proposed algorithm can satisfy the tag reading requirements specified by the upper-layer localization system with a probability as high as 0.976.