Optimal Selection of an Inertial Sensor for Cycle Slip Detection by Considering INS Output Errors

Younsil Kim, Junesol Song, Byungwoon Park, Changdon Kee

Abstract:	Currently, the m-lever of positioning accuracy is easily achieved by using pseudorange measurement of GPS satellite and is widely used in many kinds of industry already. However, for the cm-level of positioning, the pseudorange measurement is not enough due to its large noise level. For that reason, the carrier phase based positioning is necessary to achieving cm-level accuracy. The cm-level positioning has not been yet popularized as much as m-level positioning because of its complexity and high price of implementation. These days, carrier-phase based positioning is widely studied for services demanding high accuracy positioning; aircraft navigation, land vehicle and etc. However, for using carrier-phase as a measurement for positioning, carrier-phase fault detection is certainly performed before the positioning. Cycle-slip detection is one of the most important issue to implement a GPS carrier phase positioning system to a vehicle, because cycle-slip is very frequent to the weak carrier signal. It results in a degradation of the vehicle’s position accuracy with a wrong ambiguity resolution. Cycle slip detection algorithms have been studied and developed for several decades. In the capacity of GNSS receiver alone, phase-code comparison, phase-phase ionospheric residual, doppler integration, differential phases of time methods are used as a detecting technique for cycle slips. Among these methods, the only phase-phase ionospheric residual method can detection small number of cycle slips due to the large noise level of other observables; pseudorange and doppler. Recently, the several refined phase-phase ionospheric residual methods were proposed for in situation of high ionospheric activity. However, phase-phase ionospheric residual method has its own limitations about GPS receiver capacity. Only the multiple frequency receiver can utilize this method. Consequently, the system cost could be expensive. For that reason, the INS is occasionally suggested to detect the small number of cycle slips. This method can be implemented in the capacity of a single frequency receiver also. By using the INS position estimation, the next epoch of carrier phase observables are predictable. The accuracy of INS position between the consecutive epochs determines the cycle slip detection performance. Accordingly, the inertial sensor selection is critical to the optimal system design in aspect of cycle slip detection performance and the over whole system cost. Therefore, in this paper, the optimal selection of an inertial sensor for cycle slip detection is considered. The satellite-difference and time-difference residual between the predicted and measured carrier phase is defined as a monitoring value for cycle slip detection. The INS position is used for the calculation of predicted observation and its error mainly contribute to the residual. For the selection of proper inertial sensor, we conduct theoretical analysis. The INS output errors are analyzed to calculate the upper bound of estimated position error by considering the cycle slip detection performance. On the other hand, commercial inertial measurement units (IMUs) have a certain relationship for gyro and accelerometer performance to construct the IMUs. By investigating certain manufacturers, we can induce the relationship. From this result, finally we can select the proper IMU for cycle slip detection. By selecting the IMU that elicits the upper bound of INS accuracy for cycle slip detection, we can satisfy both required cycle slip performance and cost-effectiveness. Then the analytic solution is also applied to the real world. We select the appropriate IMU among the existing commercial IMUs by considering the theoretical result. Finally, the land vehicle experiment is conducted for verification of IMU selection result.
Published in:	Proceedings of the 28th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2015) September 14 - 18, 2015 Tampa Convention Center Tampa, Florida
Pages:	581 - 617
Cite this article:	Kim, Younsil, Song, Junesol, Park, Byungwoon, Kee, Changdon, "Optimal Selection of an Inertial Sensor for Cycle Slip Detection by Considering INS Output Errors," Proceedings of the 28th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2015), Tampa, Florida, September 2015, pp. 581-617.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In