Improved particle filter algorithm for INS/GPS integrated navigation system

The INS/GPS navigation system is obvious nonlinear under the large initial condition errors. To tackle the accuracy of integrated navigation under nonlinear model, a improved particle filter named cubature particle filter (CPF) is applied to INS/GPS integrated navigation. For this, the nonlinear state model based on the platform misalignment angle and the observation model described by the velocity error and position error is established. The CPF that utilizes the current observation information in the prior updating phase develops the importance proposal distribution function by cubature kalman filter (CKF) to relieve particle sample degeneracy and impoverishment, so the improved particle algorithm significantly improves the accuracy of the integrated navigation system than traditional particle filter (PF).