The pseudolite-based indoor navigation system using Ambiguity Resolution On The Fly

摘要

This paper presents a method of AROF (Ambiguity Resolution On The Fly) with extended Kalman filter (EKF) to resolve ambiguities for pseudolite-based indoor navigation system. The carrier phase measurements of pseudolites can obtain high positioning precision. In many pseudolite systems which recently have been proposed and tested, pseudolites are usually used to offer positioning and navigation applications in indoor or blocked environments. However, like indoor environment, the carrier phase measurements of pseudolite are not entirely same as GPS. A major difference is carrier phase integer ambiguity resolution. The traditional way of ambiguity resolution (AR) for GPS is static or kinematic initialization. But the static initialization is not suitable for pseudolites. Using pseudolites for static initialization, the equations are correlated with each other at observation epochs and the ambiguities can't be calculated from the equations. As a result, the integer ambiguity resolution is using the initial position of receiver as a known parameter. This paper proposes a method of ambiguity resolution on the fly (AROF) with EKF, which doesn't need to know the staring vector of receiver and achieve the kinematic initialization of carrier phase measurement in pseudolite-based indoor navigation system. As same as GPS measurement equations, Dual-Differential observation model is given based on indoor-pseudolite positioning system, which is non-linear and dynamic model. In general, the standard Kalman filter can't deal with this nonlinear situation, since the covariance equations are based on the linearized system and not the true nonlinear system. So, the best approach to this problem is to use the Extended Kalman Filter (EKF). As a result of this approach, the measurement equations of the Kalman filter become linear, and the computational requirements are significantly reduced, making it possible to estimate ambiguity in real time. Extensive testing of the filter with s--ynthetic data proved it to be satisfactory. Test cases included the presence of large initial errors as well as high noise levels. In all cases the filter was able to get ambiguities.