Multi-sensor Data Fusion for Aerodynamically Controlled Vehicle Based on FGPM

摘要

In this space race, all the countries are trying to improve the performance in every launch mission they take place. Aerodynamic as well as ground vehicles requires continuous, accurate and reliable positioning. Global Positioning System (GPS) provides most of the navigation data of the vehicles. However, signal deterioration due to ionospheric scintillation, Doppler shift, multipath of GPS lead to incorrect data. Other than GPS, inertial sensors provide position information. But each sensor has its own limitations. In order to compensate these error issues, sensor fusion using Extended Kalman Filter (EKF) is used. But due to the high dependency on INS data during GPS deterioration period makes the system less accurate. To obtain continuous and error free information from GPS, Fractional-order Grey Prediction Model (FGPM) is proposed. Grey Model (GM) requires only a limited amount of data and it predicts the GPS data during the GPS deterioration time. A GPS system along with the Fractional grey Model is be fused with the Inertial Navigation Sensor (INS) to reduce the signal losses.