Improvement of inertial surveys through post-mission network adjustment and self-calibration

Abstract

Present day inertial surveys are limited to single traverse runs in which the number of unknown system parameters to be determined are few, depending on the number of control points available along the traverse. Further, conventional inertial surveys are generally restricted to the determination of coordinates with no possibility for a rigorous post-mission adjustment of the observations. The consequence is the continued presence of systematic trends in the residuals, even after the use of error models such as those proposed by Ball, Gregerson or Kouba. Future work aiming at higher accuracies obviously requires more comprehensive models and rigorous adjustment procedures. These can be accomplished by the development of such error models and by the use of “area surveys”, instead of the single traverses, together with rigorous adjustment procedures suitable for the network of criss-crossing lines inertially surveyed. In such a network the cross-over points serve as constraints for the geodetic parameters (latitude, longitude, height, gravity anomaly, deflection components) and allow the addition of hardware and software related error parameters. Thus an opportunity is provided to effectively self-calibrate the system—a concept successfully used, for example, in photogrammetry or in satellite tracking. The number and the strength of such parameters depend on the number of control and cross-over points. The adjustment, of course, also provides the necessary statistical information on the adjusted parameters, such as their precision and the correlation between them.

The presentation will describe current work at OSU in this area.