Initial In-Mine Position Estimation Using RFID Tags

Abstract

Determining the real-time position of an underground mining vehicle relative to a global map is a very important problem facing the mining industry. Unfortunately, techniques such as Global Positioning System (GPS) cannot be used in underground mines. Alternate approaches must therefore be explored. Promising research from the field of robotics provides a possible approach to in-mine positioning, using a technique known as Simultaneous Localization and Mapping (SLAM). It is crucial that the initial position given to the SLAM algorithm is accurate. This paper investigates the concept of using an array of Radio Frequency Identification (RFID) tags placed at know positions to provide the initial position to the SLAM algorithm. A Least Squares based position estimator is presented and evaluated in an experiment. The estimator's average error is calculated for a number of cases. It is shown that the estimator presented in this paper compares well with an estimator presented previously but uses less than 8 times as many tags. The results suggest RFID based positioning, using this Least Squares approach, has the potential to provide accurate and low-cost initial position estimation.