Performance Analysis of Constrained Loosely Coupled GPSINS Integration Solutions 英文参考文献.docVIP

Sensors 2012, 12, 15983-16007; doi:10.3390/s121115983 OPEN ACCESS sensors ISSN 1424-8220 /journal/sensors Article Performance Analysis of Constrained Loosely Coupled GPS/INS Integration Solutions Gianluca Falco 1,*, Garry A. Einicke 2, John T. Malos 2 and Fabio Dovis 3 1 Istituto Superiore Mario Boella, Via P.C. Boggio 61, 10138 Torino, Italy 2 CSIRO Exploration and Mining, Pullenvale, QLD 4069, Australia; E-Mails: garry.einicke@csiro.au (G.A.E.); john.malos@csiro.au (J.T.M.) 3 Politecnico di Torino, Department of Electronics and Telecommunications, Corso Duca degli Abruzzi, 24, 10129 Torino, Italy; E-Mail: fabio.dovis@polito.it * Author to whom correspondence should be addressed; E-Mail: falco@ismb.it; Tel.: +39-011-2276-404; Fax: +39-011-2276-499. Received: 7 September 2012; in revised form: 8 November 2012 / Accepted: 9 November 2012 / Published: 20 November 2012 Abstract: The paper investigates approaches for loosely coupled GPS/INS integration. Error performance is calculated using a reference trajectory. A performance improvement can be obtained by exploiting additional map information (for example, a road boundary). A constrained solution has been developed and its performance compared with an unconstrained one. The case of GPS outages is also investigated showing how a Kalman filter that operates on the last received GPS position and velocity measurements provides a performance benefit. Results are obtained by means of simulation studies and real data. Keywords: loosely coupled integration; Kalman filter; constraints; GPS outages 1. Introduction The error in an inertial system grows very quickly over time even if when an initial calibration procedure has been performed. Position error bias on startup also significantly affects position error over time. In fact, an initial calibration can correct short term errors only and the position error can become unacceptable after a very short period of