Abstract: | The prediction of augmented GPS performance for spread-out user locations requires analyses of both accuracy under normal conditions and integrity in the case of system failures. Methods that combine covariance propagation and Monte Carlo simulation for the Wide Area Augmentation System (WAAS) have been developed, allowing system designers to study performance, risk, and cost tradeoffs. This process can be automated into computer search techniques that make WAAS network optimization possible Revisions to our previously published accuracy and integrity algorithms have been made, including more detailed WAAS accuracy models and probability models for spacecraft, ionosphere, and ground errors. Updated results are given for the FAA testbed (NSTB) network and for an example WAAS system for Europe. Next, a framework for network optimization is constructed from two bases. A top-level user value model expresses’ the relative quality of the combined accuracy and integrity evaluations for a given network. Global optimization is carried out using a genetic algorithm which maintains a population of possible network designs and “evolves” the next generation using operators derived from the Theory of Natural Selection. The optimization process is computer-intensive but has the potential to converge to the best possible network for a given application. A complete model for European WAAS network optimization is presented, and the prospects for improved computer speed using parallelized code are discussed. |
Published in: |
Proceedings of the 8th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GPS 1995) September 12 - 15, 1995 Palm Springs, CA |
Pages: | 1403 - 1415 |
Cite this article: | Pullen, Samual, Enge, Per, Parkinson, Bradford, "Global Optimization of GPS Augmentation Architectures Using Genetic Algorithms," Proceedings of the 8th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GPS 1995), Palm Springs, CA, September 1995, pp. 1403-1415. |
Full Paper: |
ION Members/Non-Members: 1 Download Credit
Sign In |