In recent years, there has been a growing interest in localization algorithms for indoor environments. In this paper, we have developed an enhanced filtering method for indoor positioning and tracking applications using a wireless sensor network. The method combines position, speed, and heading measurements with the aim of achieving more accurate position estimates both in the short and the long term. Using as a base, the well-known extended Kalman filter, we have incorporated two novel measurement covariance matrix tuning methods. The power threshold covariance matrix tuning method and the distance statistics covariance matrix tuning method, both based on the statistical characteristics of the distance estimations. In addition, we take into account the inertial measurements obtained from a nine-degrees of freedom inertial measurement unit. The system has been validated in real scenarios and results show that it provides long-term accuracy, that is, the accuracy remains below 1 m during a 20-min test. In summary, our methods benefit from the reduced observation error of the inertial sensors in the short term and extend it over a long period of time. © 2001-2012 IEEE.
- Kalman filters