Overview
In response to the Champ Announcment of Opportunity a proposal for the assessment of rapid orbit determination strategies for LEO missions has been submitted to the Champ Project Office.
The proposal aims at the analysis and validation of rapid orbit determination concepts for low Earth orbiting satellites based on GPS measurements. Both real-time algorithms for onboard applications as well as ground based processing schemes utilizing auxiliary data products for increased accuracy will be addressed within the study. The expected results are of immediate relevance for upcoming remote sensing missions demanding high accuracy orbit informa-tion in near real-time. The study will be based on TurboRogue Space Receiver 2 (TRSR-2) GPS data as well as reference orbits and gravity field models from the CHAMP science data archive. It will jointly be performed by DLR/GSOC and UNB/GGE. The study will be conducted jointly by the GPS Technology and Navigation Group of DLR's German Space Operation Center under the lead of Dr. O. Montenbruck and the research group of Prof. R. Langley at the University of New Brunswick (UNB). The team comprises the co-investigators Dr. E. Gill, Dipl. Ing. Ch. Arbinger and M.Sc. S. Bisnath.
Background
Aside from purely scientific applications like gravity field studies and meteorology, GPS is gradually evolving into a standard tracking system for low Earth orbiting (LEO) satellites. From a mission operations perspective it provides improved accuracy and coverage as well as reduced system cost in comparison to conventional ground-based S-band tracking. Fur-thermore, the onboard availability of orbit information offers the prospect of increased space-craft autonomy and a further simplification of spacecraft operations. Over the past decade, great efforts in receiver technology and data processing have been made to improve the absolute orbit determination accuracy and a decimeter level is presently achieved in the post-facto processing of spaceborne dual-frequency measurements. Aside from high accuracy, increasing attention is now paid to a near real-time orbit determination to deliver rapid end-user ready data products. Typical examples include the METOP, TerraSAR-X and SAR-Lupe missions, which demand meter-level localization accuracy and allow as little as 2 1/4 hours between data collection and dissemination of processed science or image data. The deactivation of Selective Availability in May 2000 has resulted in a pronounced simplification of the GPS-based precise orbit determination process for LEO spacecraft, which can nowadays be performed without the need of a priori GPS constellation orbit adjustment at the satellite control center. High rate GPS clock solutions are no longer required to determine the position of a user satellite, while precise post-processed orbits of the GPS constellation are readily available from various public sources. At the same time, the International GPS Ser-vice (IGS) has responded to the increasing need for a fast provision of precise GPS ephemerides by the recent installation of an ultra-rapid ephemeris service. Given the above background, the present proposal aims at the study of rapid orbit determina-tion concepts for low Earth orbiting satellites based on GPS measurements and an analysis of achievable accuracies for various classes of receivers. Both real-time algorithms suitable for onboard applications as well as ground-based processing schemes utilizing auxiliary data products for increased accuracy will be evaluated.
Objectives
The achievable orbit restitution accuracy of a LEO satellite shall be assessed under the as-sumed constraints of near-real time operation and potential hardware limitations. The results shall support the selection of adequate GPS receivers and a trade-off between accuracy and timeliness requirements in the design of future missions and operations concepts. Among others, the following key questions shall be addressed in this context:
As part of the data analysis, suitable data screening concepts and appropriate models for the deterministic and stochastic aspects of the spacecraft dynamics shall, furthermore, be worked out.
Data Analysis Methods
As a central part of the study, orbit determinations for selected CHAMP data arcs will be car-ried out using different combinations of measurement types, models, auxiliary data, and processing schemes. More specifically, the following alternatives will be evaluated:
To restrict the overall number of test cases, only a subset of suitable combinations will be treated, which is considered to be representative of real mission applications. The study builds up on previous work performed with the first public CHAMP data set of August 2000. There, focus has been given to the processing of pseudorange measurements and a technique for ionospheric correction of single frequency measurements has been as-sessed. In view of the complexity of the gravitational and non-gravitational forces acting on the CHAMP satellite at its 400-450 km altitude, promising results have been obtained with a reduced dynamics approach that will also be followed in the present study. Further work will be required to incorporate carrier phase measurements and to refine data editing concepts, dynamical models, and filter parameters. Special attention will be paid to an analysis of ob-served drag forces and its proper description by deterministic models and stochastic accelerations.
Expected Results and Application Potential
The study will provide a detailed assessment of different approaches for real-time and rapid orbit determination applications of LEO satellites. The relative merits of kinematic and re-duced-dynamic processing schemes will be compared in terms of achievable accuracy, availability of the solution and requirements for auxiliary data in the overall processing chain. Mission designers will thus benefit from a decision tree and trade-off between accuracy and timeliness requirements for orbit determination facilities of future remote sensing and science missions (e.g. METOP, TerraSAR-X, SAR-Lupe, CartWheel). For onboard applications, an independent comparison will be made that accounts for the reduced modeling capabilities faced in a real-time processing environment. Results are expected to foster an increased autonomy and onboard data processing in future European spacecraft. Optimized filtering schemes and dynamical/stochastical models will be established for both real-time and offline applications. Taking into account the limited availability of geodetic quality spaceborne GPS receivers, focus will also be given to L1-only processing schemes to avoid a restricted availability of the study results. As by-product of the data processing effort performed within the study a better understanding of the TRSR-2 receiver's clock and tracking behavior will be gained, which is expected to result in improved data screening and weighting concepts.
Further Reading
Montenbruck O., Gill E.;Ionospheric Correction for GPS Tracking of LEO Satellites;The Journal of Navigation 55, 293-304 (2002).
Montenbruck O.;Kinematic GPS Positioning of LEO Satellites using Ionosphere-free Single Frequency Measurements;Aerospace Science and Technology 7, 396-405 (2003).DOI 10.1016/S1270-9638(03)00034-8
Montenbruck O., Kroes R.;In-flight Performance Analysis of the CHAMP BlackJack Receiver;GPS Solutions 7, 74-86 (2003).DOI 10.1007/S10291-003-0055-5
Gill E., Montenbruck O.;Comparison of GPS-based Orbit Determination Strategies;18th International Symposium on Space Flight Dynamics, 11-15 Oct. 2004, Munich, Germany (2004).
Montenbruck O., Gill E., Kroes R.;Rapid Orbit Determination of LEO Satellites using IGS Clock and Ephemeris Products;GPS Solutions (2005).DOI 10.1007/s10291-005-0131-0
Montenbruck O., van Helleputte T., Kroes R., Gill E.;Reduced Dynamic Orbit Determination using GPS Code and Carrier Measurements;Aerospace Science and Technology 9/3, 261-271 (2005).DOI 10.1016/j.ast.2005.01.003