Delayed fusion for real-time vision-aided inertial navigation

In this paper, we consider the effects of delay caused by real-time image acquisition and feature tracking in a previously documented vision-augmented inertial navigation system. At first, the paper illustrates how delay caused by image processing, if not explicitly taken into account, can lead to appreciable performance degradation of the estimator. Next, three different existing methods of delayed fusion and a novel combined one are considered and compared. Simulations and Monte Carlo analyses are used to assess the estimation errors and computational effort of the various methods. Finally, a best performing formulation is identified that properly handles the fusion of delayed measurements in the estimator without increasing the time burden of the filter.