Nonlinear Autoregressive Neural Network Integrated Approach for Tackling GPS Outages Intervals

Nowadays inertial navigation system and global positioning system (INS/GPS) integrated systems are widely used in many moving vehicles. The Extended Kalman Filter (EKF) is utilized extensively in the integration task to obtain accurate and continuous estimated position. However, the efficiency of EKF to estimate position degrades in case of GPS signal outages due to signal blockage or jamming. In this work a Nonlinear Autoregressive Neural Network with Exogenous input (NARX) based technique is proposed to estimate the position during GPS outages Two simulation environments are created for both ground and aerial vehicles alongside three inertial sensors with different accuracy grades to validate the proposed algorithm. The conducted experiments reveal the superiority of the proposed method in estimating accurate and continuous position at different GPS outage’s intervals.