INTRODUCTION The integration of GPS receivers with inertial navigation systems (INS) has been well investigat- ed in the past. The advantages of integrated sys- tems relative to GPS-only are reported to be a full 6 degrees of freedom navigation solution, improved accuracy, smoother trajectories, and reduced sus- ceptibility to jamming and interference [1]. Other studies have examined the benefit of using the inertial solution to improve ambiguity resolution performance [2 – 4]. The extent to which the above advantages hold is well known to be dependent on the quality of the inertial measurement unit (IMU) used in the inte- gration. With this in mind, it is not surprising that past investigations have typically looked only at using high-end, navigation-grade inertial systems for high-accuracy (centimeter-level) applications. Unfortunately, the cost of these inertial units can be prohibitive for many applications and/or agencies. With the decreasing cost of tactical-grade inertial sensors, however, their use in integrated navigation systems is desirable. Yet the question arises of whether these lower-cost units will still be of bene- fit for applications demanding the highest accura- cies. From a different standpoint, could the cost of a tactical-grade IMU be justified by the resulting improvements in the integrated solution relative to the GPS-only case? This paper aims to quantify the benefits of using a tactical-grade IMU in high-accuracy navigation systems compared with using GPS alone. This work is novel in light of previous investigations that have used primarily navigation-grade units to obtain the highest navigation accuracy. In particular, the fol- lowing performance parameters are investigated: Position accuracy during complete and partial GPS data outages. Results of this investigation determine the operational conditions under which the system can be used to maintain a given level of performance. Positioning performance is assessed for the situation in which L1 or wide- lane carrier-phase ambiguities are available. Velocity accuracy is also investigated for applica- tions to which this parameter is most relevant. The time needed to determine the integer ambiguities after complete and partial GPS data outages. The veracity of the ambiguity fix is also considered. Again, the investigation addresses applications requiring L1 or wide- lane ambiguities. The above parameters are investigated by simulat- ing GPS data outages during postmission processing of real field data. A reference trajectory obtained using all available data is used for comparison purposes to assess system accuracy during the data outages. After the simulated data outages, the time required for the system to recover the fixed ambiguities is measured, as is the accuracy of the ambiguity fix. In addition, by cross-referencing the ambiguity res- olution performance with the positioning accuracy 1 Benefits of Using a Tactical-Grade IMU for High-Accuracy Positioning M. G. PETOVELLO, M. E. CANNON, and G. LACHAPELLE University of Calgary, Calgary, Alberta, Canada Received June 2003; Revised October 2003 ABSTRACT: Integration of GPS with inertial sensors can provide many benefits for navigation, from improved accuracy to increased reliability.The extent of such benefits, however, is typically a function of the quality of the inertial system used. Traditionally, high-cost, navigation-grade inertial measurement units (IMUs) have been used to obtain the highest position and velocity accuracies. However, the work documented in this paper uses a Honeywell HG-1700 IMU (1 deg/h) to assess the benefits of a tactical-grade IMU in aiding GPS for high-accuracy (centimeter-level) applications.To this end, the position and velocity accuracy of the integrated system during com- plete and partial GPS data outages is investigated. The benefit of using inertial data to improve the ambiguity resolution process after such data outages is also addressed in detail. Centralized and decentralized filtering strategies are compared in terms of system performance. NAVIGATION: Journal of The Institute of Navigation Vol. 51, No. 1, Spring 2004 Printed in the U.S.A.