Maximum likelihood estimation for orbit determination
Abstract
Advanced estimation techniques for orbit determination and prediction have been investigated, and the results are reported. The baseline least squares estimation (LSE) orbit determination algorithm that was used as a reference for all performance comparisons is described. Then, the maximum likelihood estimation (MLE) orbit determination algorithm that was utilized is described, together with alternative measurement frames and candidate disturbance error models. A comparison of numerical results for the baseline LSE and the improved MLE algorithms is provided for five real-world test orbits covering a range of orbital altitudes and eccentricities. Two of these test cases show a sixfold improvement in orbit prediction accuracy due to MLE, two show a twofold improvement, and one shows no improvement over the baseline LSE algorithm.
- Publication:
-
Astrodynamics 1985
- Pub Date:
- August 1986
- Bibcode:
- 1986asdy.conf..273B
- Keywords:
-
- Maximum Likelihood Estimates;
- Orbit Calculation;
- Accuracy;
- Covariance;
- Error Analysis;
- Kalman Filters;
- Least Squares Method;
- Astrodynamics