Sımu ve Gnss Ölçülerinin Entegrasyonuna Dayalı Mobil Yersel Gravimetri

Abstract

The Earth’s gravity field can be measured with a certain resolution and accuracy using absolute and relative gravimeters installed on a stationary (static) or moving (mobile) base ground, airborne and spaceborne platform. Each gravimetry method has its own advantages and disadvantages in terms of observable signal magnitude, resolution, coverage, and cost. Static terrestrial gravimetry performed with free-fall absolute gravimeters and spring-based relative gravimeters is a far superior technique in terms of accuracy and magnitude of the observable signal, but it is time consuming compared to mobile and band-limited airborne, seaborne, and satellite gravimetry methods. Terrestrial gravimetry contains the full spectrum of the Earth’s gravity field i.e., short, medium, and long-wavelength information due to the spatial proximity to the masses causing gravitation. The moving observation platform used in the mobile airborne, seaborne and satellite gravimetry can acquire data with a particular speed and from a certain height (except for the sea), the corresponding gravity field solutions are adversely affected by relatively high speeds of the moving platforms as well as the ill-posed downward continuation operation (except the sea). The mobile airborne and shipborne gravimetry lie between the terrestrial and satellite gravimetry and aim to compensate for the weaknesses of both. Satellite gravimetry has a global coverage and provides long-wavelength gravity field information. Theoretically, it is possible to collect gravity data on the Earth surface using a mobile gravimetric system to be placed on a land vehicle with lower speed and without any interruption. This may provide quite similar precision to classical terrestrial techniques but with much higher resolution and shorter data collection duration. In the thesis, a mobile terrestrial gravimetry system prototype has been developed that can reveal these potential advantages. The developed system based on the integration of Strapdown Inertial Measurement Unit (SIMU) and Global Navigation Satellite Systems (GNSS) has been tested on a 45-km long route consisting of 23 ground control points in Ankara province. The SIMU/GNSS integration is implemented using a loosely-coupled closed-loop Extended Kalman Filter (EKF) with 18 state vectors. The vertical gravity disturbance is augmented to the system state vector as a stochastic process similar to the inertial sensors’ errors. The terrestrial static gravity observations and zero-velocities at the start and end points along with GNSS position and velocity solutions have been introduced to EKF as observations. EKF solutions in the forward direction in time have been smoothed by the Rauch–Tung–Striebel (RTS) method. The closure errors have been determined by comparing the gravity disturbance estimates from mobile gravimetry with the corresponding high-precision gravity disturbance values at the ground control points. The preliminary results show that vertical gravity disturbance can be obtained with a mean accuracy of 4.2 ± 2.1 mGal using terrestrial mobile gravimetry method. The non-availability of zero-velocity and gravity measurement updates at intermediate control points between the departure and arrival, GNSS outages in the urban areas, unexpected problem in the SIMU temperature stabilization, stochastic models and EKF parameters used in the analysis can be listed as the possible causes of the closure errors. It is possible to say that the differences are slightly higher especially at the control points located in the urban areas where the GNSS outages are experienced. The internal temperature of the SIMU has changed by about 8°C from the beginning to the end of data collection, which may cause drift in the gravity solutions due to the temperature-dependent accelerometer bias. Moreover, the EKF estimations of SIMU sensor errors grow rapidly since the GNSS measurement updates have not been applied due to signal outages which may affect the gravity solutions. It is planned to repeat the test after fixing up the problem in the SIMU temperature stabilization system, improve the analysis software used in the study in order to implement zero-velocity and gravity measurement updates at intermediate points, test different stochastic models and tune EKF parameters used in the analysis, and integrate additional sensors such as odometer and multi-antenna GNSS receiver.