New, Generally Applicable Metrics for RAIM/AAIM Integrity Monitoring

P.B. Ober

Abstract:	In this paper, we will develop a new integrity metric for navigation systems, that we will call the Bias Integrity Threat (BIT), and a derivative of it, the Maximum Undetectable Position Bias (MUPB). Integrity metrics are used to indicate whether a navigation system has sufXcient error detection power. As we will see, integrity depends on both measurement geometry and measurement quality, that is, the standard deviation of the measurement noise. Existing metrics like the &f- [Sturza90] and the ARP [Brown92b] have two serious limitations: they can not be applied straightforwardly for measurements of different standard deviations, or for multiple failures. Also, they can not be used in a KaIman filter environment. Because these metrics do not take the noise level into account, the values of L$$- and the ARP are relative, rather than absolute measures of integrity, which makes it hard to interpret their values. We will illustrate this in section 7. The BIT and MUPB do not possess any of these disadvantages: they can be used with general measurement covariance matrices, and are well defined for the assumption of multiple failures. They are therefore applicable in a wide range of applications, including Kalman filter environments. Furthermore, their values are absolute and thus more suitable for comparing the integrity of different navigation systems than the traditional integrity measures. Their computational needs are modest, allowing their use in real time applications.
Published in:	Proceedings of the 9th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GPS 1996) September 17 - 20, 1996 Kansas City, MO
Pages:	1677 - 1686
Cite this article:	Ober, P.B., "New, Generally Applicable Metrics for RAIM/AAIM Integrity Monitoring," Proceedings of the 9th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GPS 1996), Kansas City, MO, September 1996, pp. 1677-1686.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In