Abstract: | The estimation accuracy of a conventional Kalman filter is ultimately limited by imperfect mathematical modeling of the input noise and this issue has become a barrier to improving estimation performance for some applications. In contrast, the wave estimation method, which describes the dynamic system in a deterministic way over a short period, has the advantages of providing higher estimation accuracy in a situation where the input disturbances are of low frequency, slowly varying in nature, and rely on relatively weak observables. In this paper, a wave estimator has been developed to integrate differential GPS (DGPS) with a medium accuracy Honeywell HG1700 INS, for land positioning with the emphasis on accuracy at the meter level. A loosely coupled integration approach has been developed, which uses carrier phase-smoothed C/A-code DGPS positions and velocities as updates to the INS. A field test was carried out whereby raw data from two NovAtel OEM-3 receivers were logged at 1 Hz rate; and the raw angular velocities as well as specific force measurements from the INS were logged at 100Hz. A reference trajectory was obtained by using the carrier phase kinematic positioning technique. The dataset was processed using a conventional Kalman filter as well as the wave estimator. The results and related analysis are presented, which shows an improvement of estimation performance by using the wave estimation technique. |
Published in: |
Proceedings of the 58th Annual Meeting of The Institute of Navigation and CIGTF 21st Guidance Test Symposium (2002) June 24 - 26, 2002 Hyatt Regency Hotel Albuquerque, NM |
Pages: | 241 - 249 |
Cite this article: | Zhang, X., Cannon, M.E., "Integration of GPS with a Medium Accuracy INS for Meter-Level Positioning Using a Wave Estimator," Proceedings of the 58th Annual Meeting of The Institute of Navigation and CIGTF 21st Guidance Test Symposium (2002), Albuquerque, NM, June 2002, pp. 241-249. |
Full Paper: |
ION Members/Non-Members: 1 Download Credit
Sign In |