A data fusion method based on navigation observations from multiple sensors with asynchronous sampling rates is presented aiming at the accomplishment of an optimal navigation result. Firstly, the observation estimations of navigation sensors at the same designated time are obtained through the fitting or interpolation algorithms based on the multi-sensors’ measurements at different times, which solves the observation inconsistency resulting from asynchronous sampling frequencies of multi-sensors. Secondly, the optimal navigation result is achieved through assigning appropriate weights to the above observations or estimations based on the relationships among them. Finally, the data fusion method is validated by simulation, and the results demonstrate that it can effectively reduce the observation noises of multi-sensors and improve the accuracy of navigation.