, n-pariz@ferdowsi.um.ac.ir
Abstract: (10228 Views)
The estimation of situation in a combinational navigation GPS/INS with least number of satellites is the main purpose of this paper. As inertial measurement unit uses altimeter for height measurement, we can assume which height poses certain amounts, whereas geographical length and width are unknown to us in this paper. The single difference GPS is employed for updating the inertial navigation system and also there is a proper use of particle filter for filtering and error estimation in this system, although regarding to high dimension of state equations in combinational navigation GPS/INS the use of particle filter needs a lot of calculations and consequently more time. As the kalman filter could be the optimum filter for the linear systems which bear gussian noise, the use of kalman filter for estimation is recommended for linear equation section and use of particle filter in nonlinear section. The comparison of mentioned algorithm's results with extended kalman filter, indicated when there is two GPS satellites the error estimation of situation is bounded by using this method, whereas the application of extended kalman filter increases the error.