Improvement of Positioning Accuracy of Landslide Monitoring Systems by GPS

Shinji Oda, Mikio Nakamura, Katsuo Yui, Masahiro Genda, Hiroaki Toyoizumi, Tatsuji Fujise, Yasushi Watanabe and Jun Sato

Abstract:	GPS has been employed in various applications including disaster prevention systems in Japan, such as to monitor landslide. Among the requirement for disaster prevention systems, GPS is considered to be an appropriate technique for landslide monitoring systems because of its three dimensional positioning feature. A landslide monitoring system is a disaster prevention system to estimate the hazard level of landslide area by monitoring the ground displacement in millimeters over a long term. Since a landslide monitoring system, in general, is installed in a mountainous region, the installation condition should be carefully selected in advance. For example, the cliffs or the woods in the vicinity of the system cause multipath interference, which disturbs the monitoring system from detecting an abrupt ground displacement, while such detection is essential for landslide monitoring. For this reason, the installation condition for conventional systems should be carefully chosen not only to retrieve satellite signals, but also to avoid measurement error caused by multipath such that a favorable open view as possible for the system; however, the multipath interference in a mountainous region is not sufficiently suppressed by merely displacing the system. We have developed the techniques to improve measurement accuracy and to achieve detection of abrupt displacement. We evaluated the developed techniques by applying them to our system. The techniques we have developed are summarized as follows: (1) Measurement Accuracy Measurement accuracy is deteriorated mainly by multipath error, which includes various factors. We applied a statistical method to suppress multipath error, improving the accuracy. (2) Detection of Abrupt Displacement In the conventional systems, the detection response to a rapid displacement tends to be delayed, because of the low-pass filter to suppress multipath error. We have simultaneously, whereas abrupt displacement of tens of mm/hour scale, and long-term displacement of mm/day scale should be detected. We propose a hybrid filter by combining a Kalman filter and a displacement estimation filter for solving the problem of conventional methods, which is specialized for landslide monitoring systems. As in Figure 1, the hybrid filter consists of three functional blocks: the Kalman filter, the displacement estimation filter, and the filter adjustment process. The Kalman filter improves the positioning accuracy according to an observation model we propose later in this paper. The displacement estimation filter is used to calculate the amount of
Published in:	Proceedings of the 17th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2004) September 21 - 24, 2004 Long Beach Convention Center Long Beach, CA
Pages:	261 - 267
Cite this article:	Oda, Shinji, Nakamura, Mikio, Yui, Katsuo, Genda, Masahiro, Toyoizumi, Hiroaki, Fujise, Tatsuji, Watanabe, Yasushi, Sato, Jun, "Improvement of Positioning Accuracy of Landslide Monitoring Systems by GPS," Proceedings of the 17th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2004), Long Beach, CA, September 2004, pp. 261-267.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In