ION GNSS 2009
Session E3: Urban & Indoor Navigation Technology 1

Title: Cost-effective Selection of Inertial Sensor for Cycle-slip Detection for Land Vehicle
Author(s): Y. Kim, J. Song, B. Park, H. Yun and C. Kee, Seoul National University, South Korea
Room: Chatham Ballroom C

This paper represents optimal inertial sensor performance selection for inertial sensor-aided cycle-slip detection for land vehicle. The inertial sensor is used to detect cycle-slip. By using the inertial sensor output, we estimate the future position of the vehicle. And cycle-slip is detected by comparing carrier phase measurement from GPS satellite with estimated range from INS position. We induce the optimal sensor performance for desired minimum detectable cycle-slip through RMS error analysis of residual according to sensor performance. And we analyze the cycle-slip detection performance by statistical analysis and GPS, inertial sensor simulation data. The m-lever positioning accuracy is achieved by using pseudorange measurement of GPS satellite. And achieving cm-level positioning accuracy, carrier-phase based positioning method is needed, such as RTK. 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 position. And cycle-slip is a one of the fault of signal. When signal reception by a GPS receiver interrupted by the shadowing or during a severe signal disturbance such as due to Ionospheric activity and multipath interference or low SNR, signal tracking interruption occurs. After the signal tracking interruption, the ambiguity has a different value than before; cycle slips. And carrier phase based positioning accuracy can be worse because of cycle slip, therefore cycle slip detection should be conducted before using the carrier phase measurement. In South Korea, there is an on-going R&D project whose goal is to prevent car accidents involving vehicles carrying hazardous materials by monitoring their positions which are estimated by RTK, whether they deviate the designated lane or not. The project is supervised by Korea Aerospace Research Institute and many universities are participating including Seoul National University. Development of cycle slip detection algorithm is included in the project, and we devote to it. Cycle slip detection algorithms have been studied and developed for several decades. 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, phase-phase ionospheric residual method is usually used for detecting cycle slips because of this method is easy to implement and can be utilized both static and dynamic users. However, phase-phase ionospheric residual algorithm is based on the dual frequency receiver. It is not cost effective compare to single-frequency receiver. Cost effectiveness is very important to our goal because the vehicle that carrying hazardous materials are so many, we cannot equip expensive equipments for every vehicles. Furthermore, phase-phase ionospheric residual method cannot detect some special cycle slip cases where the ratio of L1 and L2 cycle slip is close to the ratio of wavelengths of L2 and L2 frequencies. It degrades the reliability of positioning. To improve these limitations, we proposed the algorithm which integrating GPS and low-cost IMU. It is cost effective compare to dual frequency receiver. And we also study the optimal sensor performance for our cycle slip detection algorithm to select the profit IMU among many IMUs in the market they have different performances and different costs. This paper focuses on the optimal inertial sensor performance selection for detecting the desired minimum cycle slips, and cycle slip detection algorithm. Cycle slip is detect by comparing carrier phase measurement from satellite and estimated range by using INS position. And to select the profit IMU performance we induce the optimal IMU performance by using theoretical approach. We investigate the relationship between performance and cost of the commercial IMUs and establish the mathematical model for their relationship. And we induce the optimal sensor performance by solving optimal problem. Selected sensor performance is examined by using simulation data. And we analyze the cycle slip detection performance according to selected IMU. The combination of L1 and L2 measurements is not involved in the proposed algorithm, so there are no special shadowing cycle slip sets which earlier algorithms have. And we don´t use receiver difference on the carrier-phase measurement, so proposed algorithm can be applied to single vehicle independently without using data from reference station. Therefore, this algorithm can be used for Compact RTK protocol which broadcast only correction value, not the whole raw measurement. The proposed algorithm can be used to many land vehicle navigation applications such as monitoring vehicle carrying hazardous materials and autonomous driving.

Previous Abstract Return to Session E3 Next Abstract