RFID-based localization system for mobile robot with Markov Chain Monte Carlo

This paper proposes a robust and precise localization system for mobile robots with the aid of Radio Frequency Identification (RFID) technology and the Markov Chain Monte Carlo (MCMC) optimization algorithm. By integrating the RFID module, RFID tags, a mobile robot and an off-board computer together, the localization system uses MCMC to analyze RFID signal information and get optimized mobile robot´s location information. This paper focuses on explaining the development of RFID sensor model and the processes of building the MCMC algorithm. For sensor model development, the effects of orientation between the RFID reader and RFID tag and the velocity of the mobile robot are investigated. For the MCMC method, a sequential Monte Carlo method is adopted to optimize the result. Finally, the initial experiment results indicate that the proposed localization system has low error and could be used to effectively locate mobile robots.