High Precision Time Synchronization Strategy for Low-Cost Embedded GNSS/MEMS-IMU Integrated Navigation Module

Time synchronization is the key technology of real-time integrated navigation, and its error decides the precision of integrated navigation. Global navigation satellite system (GNSS) and inertial navigation system (INS) real-time integrated navigation requires the fusion of GNSS and inertial measurement unit (IMU) data at 1PPS. However, due to data computation and circuit delay, it is impossible to receive two data simultaneously at 1PPS, resulting in the inability to achieve high-precision data fusion. In response to this issue, this paper proposes a novel time synchronization strategy, which first saves IMU data, waits for GNSS data to be received, and then fuses the two. The key to this approach is to ensure that the saved IMU data is processed within a sampling interval. This article adopts a one-step predictive Kalman filter algorithm to place the state prediction covariance matrix in the traditional algorithm into the measurement update process for execution so that only INS mechanization algorithms are executed during the state prediction process, which can significantly reduce the code runtime and ensure that the saved IMU data can be processed promptly. The correctness of the proposed algorithm was verified through real-time vehicle experiments in the real world. The test results show that data time synchronization can be achieved accurately to the order of microseconds with the proposed synchronization approach. The integrated navigation system with this strategy achieves performance with comparable real-time positioning and post-processing positioning accuracy.