Autonomous orbit determination using GPS
Abstract
State estimation procedures for satellite onboard orbit determination are developed and presented. In face of restrictions imposed by the onboard computer, simple analytical models for the satellite dynamics are used in conjunction with a Kalman filter for real time processing of Global Positioning System (GPS) observations. The satellite motion is either represented by polynomial expressions that approximate the behavior in time of the orbital elements or by a harmonic oscillator that has an analytical solution thus avoiding a numerical integration. The GPS observations are then sequentially processed to identify the coefficients of the polynomials (dynamic model) and evaluate the orbit or to determine directly the satellite position and velocity independent of ground support. With the proposed procedures the use of a more complex satellite dynamical model is avoided and the numerical integration, needed in standard orbit estimation, to propagate the orbit and the state error covariance matrix. The results of the numerical tests conducted under simulated conditions in a digital computer indicate that the proposed procedures are candidates for use in autonomous systems for real time orbit determination.
 Publication:

Ph.D. Thesis
 Pub Date:
 June 1989
 Bibcode:
 1989PhDT........13N
 Keywords:

 Autonomy;
 Computerized Simulation;
 Dynamic Models;
 Error Analysis;
 Global Positioning System;
 Mathematical Models;
 Numerical Integration;
 Orbit Calculation;
 State Estimation;
 Airborne/Spaceborne Computers;
 Digital Computers;
 Kalman Filters;
 Matrices (Mathematics);
 Orbital Elements;
 Polynomials;
 Real Time Operation;
 Spacecraft Models;
 Time Measurement;
 Astrodynamics