MEMS 3D DR/GPS Integrated System for Land Vehicle Application Robust to GPS Outages

Abstract

In this paper, we propose a MEMS 3D DR/GPS integrated system which provides 3D attitude and position information for land vehicle application. Since low-cost MEMS-IMU provides data with large bias, it is hard to obtain an accurate position of the land vehicle, especially during GPS outages. To improve the performance of the MEMS-based navigation system, odometer-based 3D DR/GPS formulation is proposed. The proposed algorithm uses 1D speed information for obtaining 3D position and extended Kalman filter (EKF) to compensate for navigation errors. An error model for the EKF is developed, including the odometer scale factor, which may change over various environments and require an accurate value. By substituting a MEMS INS/GPS system to 3D DR/GPS system, the velocity and position errors of navigation solution are reduced. Since the proposed system estimates 3D attitude and position unlike conventional DR/GPS system, it suppresses navigation errors effectively when a land vehicle goes uphill and downhill even for the GPS outages. Besides, the adaptation rule for measurement noise covariance is applied to improve navigation performance. The proposed method is tested for a land vehicle in an urban area, including GPS outages. The experimental results show that the proposed MEMS 3D DR/GPS system provides a more stable navigation solution with fewer position errors.