Gravity compensation methods for precision INS

Precision inertial navigation depends not only on the quality of the inertial sensors (accelerometers and gyros), but also on the ability to compensate for the gravitational effect. With a view toward the next-generation inertial navigation system, based on sensors whose errors contribute as little as a few meters per hour to the navigation error budget, there are two viable options for gravity compensation, each with its own unique advantages and disadvantages. The first option, traditional method, relies on the existence of gravity maps based on gravity surveys provided by static ground instrumentation and airborne mapping missions. This case depends on extensive modeling using potential theory to predict the gravity vector at altitude. The second option is to determine the gravitational compensation en route (on the fly) using gravitational gradiometers that sense the curvature of the potential field. This case depends on augmenting the INS with sensitive instrumentation that to date has a long history of technology development but very little operational experience. This paper overviews the theoretical background and analyzed the effect of gravity compensation through simulation especially focused on the traditional method. In addition, various cases changing the data grid and noise level were tested and its effect was investigated using direct interpolation and least square collocation. The effects of gravity errors are presented in terms of the horizontal errors in the navigation solution so that the necessary accuracy of the gravity models can be found with given accuracy of the navigation solutions.