Abstract: Time-delay is an important issue in multi sensor data fusion. As a common positioning method of an intelligent vehicle, global positioning system/inertial navigation system (GPS/INS) always suffers from an internal time-delay which can cause an error estimation of the INS error state. Based on the traditional loosely coupled integrated navigation system, this paper proposed an algorithm for delay estimation and compensation. Firstly, the delay estimation model was formulated to estimate the time synchronization error. Then, the residual propagation equation was constructed, and the delay compensation was carried out by means of residual reconstruction. The time synchronization was implemented in a software-based scheme. According to the analysis of errors and root mean square errors of velocity and position in the experiment, the improved filtering algorithm was able to effectively reduce the error caused by GPS sampling delay and improve positioning and velocity measurement accuracy.

Key words: delay estimation; delay compensation; integrated navigation; residual reconstruction