Improvement of an MMS trajectory, in presence of GPS outage, using virtual positions

This paper show the main results obtained by some experimentations performed with the aim to improve the trajectory of the vehicle during GPS outages using GPS virtual positions. As well known the navigation system is generally constituted by one or more geodetic GNSS receivers and other sensors such as Inertial Measurement Units (IMU) and Odometers (ODO). The observations acquired by these sensors can be fused using Kalman Filter applied with different approaches (Tightly or Loosely Coupled); when all the navigation systems work well, the final accuracy can be at centimeters level. As well known in historical city centers the sky visibility can be reduced for the presence of high buildings closed to the streets. Working in wide areas the solution obtained by the data fusion between GPS and IMU sensors can be assumed as good and can constitute the reference trajectory useful to compare other solutions obtained in different ways. Starting from this reference solution other solution can be obtained performing a GPS outage simulation obtained cutting GPS solution for a defined window time. The new GPS file has been combined again with the IMU into the Kalman filter obtaining a solution in presence of a GPS outage. Starting from this last solution and disposing of the original file of GPS positions (without GPS outages), some simulation has been performed reintroducing some original GPS solutions into the file characterized by GPS outage and reprocessing the data into the Kalman filter. The reintroduction of the GPS positions simulates the correction potentially derived by topographic survey. In order to evaluate the influence of this approach in terms of accuracy and cost/benefit ratio, many tests using the three test areas has been performed and the obtained results are presented.