Search

In order to simplify the MEMS-IMU/GNSS integrated navigation system model, improve the navigation accuracy of the carrier under the motion condition of large Angle or frequent steering turning, two navigation modes are adopted. When GNSS signals are available, horizontal velocity and height error are used to construct three-dimensional measurement equations for partial integrated navigation. When GNSS signals are unavailable, kinematic constraint equations are introduced to assist MEMS-IMU for auxiliary navigation. Considering the problem that MEMS gyroscope zero is too large and compass effect is difficult to be reflected, a partial attitude feedback strategy is adopted for the feedback correction. The experimental results show that when GNSS signals are available, compared with the six-dimensional integrated algorithm of speed and position, the partial integrated algorithm simplifies the system model, reduces the dimension of the system, and reduces the amount of calculation, it also has better accuracy. When GNSS signals are unavailable, kinematics constraints auxiliary navigation can effectively restrain position offset, reduce navigation errors and improve navigation accuracy.

In order to estimate the misalignment angle of align system effectively, this paper constructs a traditional 14 state rapid transfer alignment model, considers the problem that traditional velocity error definition is not completely accord with the consistency of coordinate system, the improved differential equation of velocity error is applied to transfer alignment, constructs an improved 12 state rapid transfer alignment model. Finally, the performance of the two alignment algorithms is tested by vehicle-mounted experimental data under the conditions of no maneuver, C shape maneuver and S shape maneuver, and the experimental result of the two algorithms is compared and analyzed. The experimental result shows that the improved 12 state rapid transfer alignment algorithm not only has better alignment accuracy than the traditional 14 state rapid transfer alignment algorithm, but also reduces the model complexity, reduces the system calculation, and avoids the influence of flexural deflection and dynamic misalignment angle. The yaw angle alignment accuracy is improved by 45%, 65% and 53% under the conditions of no maneuver, C shape maneuver and S shape maneuver, respectively.