Strapdown Navigation Technology: A Literature Survey

Abstract

S inertial navigation depends on the integration of acceleration with respect to a Newtonian reference frame. Conventional systems physically maintain such a reference frame by means of gimbals. This is termed a stabilized platform, and the system gyros and accelerometers are mounted on it. Many high-performance systems of this type have evolved during the past two decades. The version of this idea, wherein sensors are mounted directly on the vehicle and the transformation from the sensor to inertial reference frame is computed rather than mechanized, has been recognized as long as that of the stabilized platform. Potential advantages of the strapdown concept, compared to stabilized platforms, are often discussed and include lower cost, reduced weight and power consumption, increased reliability, and ease of maintenance, manufacture, and redundant design. Strapdown system development commenced in the late 1950's. The first flight-operational hardware, built by Honeywell for the PRIME re-entry vehicle, was flown in December 1966. A major milestone was reached in 1969 when the Apollo abort guidance system achieved man-rated operational status. These early strapdown systems, however, operated in a relatively benign dynamic environment. Dealing with rapid maneuvers and vibrations would have required sensors with much greater dynamic range and so much onboard computation as to be prohibitive with the computers available then. In the last decade, major advances in computer technology have prompted extensive development of sensors uniquely suited to strapdown requirements. Thus a trend toward widespread use of strapdown inertial navigation has been discernible for some time. Whether strapdown navigation has arrived, however, is still a point of lively debate, cf. the 1976 papers by Shaw* and Peterson. The present paper does not enter into this debate. The benefits of the strapdown approach accrue from more than just the deletion of gimbals. The integration of inertial with other types of navigation information is facilitated, and modular design is encouraged. Since a more powerful computer is used, sophisticated data processing, estimation, and failure processing methods can be employed. Redundancy for increased reliability can be implemented conveniently at the sensor and component level, rather than at the system level, and skewed configurations can be used to minimize the number of sensors. Above all, a strapdown system has the potential to show a considerable reduction in ownership and life-cycle cost over its gimbaled counterpart. These and other advantages have been touted widely for many years; however, the development of truly competitive strapdown systems has been only quite recent. As strapdown navigation technology has matured over the years, a significant body of literature covering all aspects of the field has become available. Many review papers discuss the development of strapdown navigation in its various stages. The evolution of strapdown navigation from the conceptual stage to the development of early flight hardware is discussed in Refs. 3-11; the maturing of the strapdown approach as hardware limitations were surmounted is discussed in Refs. 12-16; and the dynamic and competitive present-day situation is described in Refs. 1, 2, 17, and 18. To the authors' knowledge, no general literature survey of the field has appeared in the past, however, and the present paper attempts to fill this role.