In complex scene，the fault of the GNSS lies in the susceptibility to be interfered and the large fluctuant positioning error, so tightly coupled IMU/EGNOS sensor fusion technique is used. To improve the positioning accuracy, an IMU/GNSS error model is established by using extended Kalman filter. By adopting IMU assisted EGNOS positioning, the adaptability of EGNOS is enhanced. The test results showed that this method greatly improved the positioning stability and accuracy of the tightly coupled IMU/EGNOS sensor fusion system.