A Novel BPNN-Based Method to Overcome the GPS Outages for INS/GPS System

Abstract

In order to improve the performance of the global positioning system (GPS) and inertial navigation system (INS) integrated system during GPS outages, a novel fusion algorithm based on back propagation neural network (BPNN) is proposed. A new model is built which relates the INS velocity, inertial measurement unit (IMU) outputs and the duration of GPS outages to the GPS position increment. Performance of the proposed method has been experimentally evaluated in a land vehicle navigation test. The test results show: (1) the proposed model can efficiently predict the increment of position and compensate the INS errors accumulation during GPS outage; (2) the advantage of new model on positioning accuracy becomes more obvious when the GPS observations are unavailable for a long time; (3) utilizing the current and past 2-step information as the input of BPNN model can effectively balance the computation burden and accuracy.