Trade-off on different autonomous orbit-determination solutions including GNSS and optical-based navigation, and a trade-off on different station keeping strategies, to define the global autonomous system best suited for chemical and for hybrid chemical/electrical Telecom platforms on geo-stationary orbit. The chosen solution is based on GNSS and on classical Station Keeping strategy. Performances are assessed on a closed-loop functional software simulator, and on a closed loop real-time test bench including a MOSAIC® GNSS receiver, a LEON2 on-board computer, a SPIRENT GNSS stimulator, and a REDHAWK PC.
The objective of this activity is to develop and evaluate the performances of on-board algorithms for the planning and execution of station keeping manoeuvres on geo-stationary orbit for fully chemical and for hybrid chemical / electrical telecommunication platforms, using autonomous orbit determination based on optical navigation or on GNSS. It is a natural follow-on to two previous ESA studies: “Feasibility of GNSS sensors for AOCS Applications in GEO and Higher Altitudes” and “GEO Orbit Determination Using an APS-Based Navigation Sensor”.
Autonomous orbit determination trade-off shown that pure orbit propagation and optical-based navigation are not sufficient for autonomous OD/SK requirements because of their insufficient accuracy (>0.02°) and because they require frequent calibrations. MOSAIC performance is sufficient to meet these requirements in nominal conditions, but better accuracy in radial position is required in degraded conditions. LION performances are expected to be sufficient in any conditions.
The chosen Station keeping strategy is similar to the one currently performed on ground because it offers a good ephemeris and manoeuvers prediction for at least the upcoming week, it has no overconsumption, and is well mastered by the operators.
Station keeping manoeuvres management from ground represents a significant work load for operators:
- Ranging campaign used for orbit determination
- Manoeuvres planning, implementation, execution monitoring, calibration
- Especially for electrical propulsion satellites (400 man / year, with man duration = 2h)
- Especially for large satellite fleet
To handle this situation, most operators have developed automatic layers to flight dynamics ground SW. Autonomous Orbit Determination and Station Keeping for Competitive Telecom platforms study is an extension of this logic.
Main expectations include:
- Better Orbit measurement availability
- Faster identification of manoeuvres errors
- More information to support anomaly investigation
- Lightened Operations, more added-value work
The system consists in two main parts:
- The navigation system: either MOSAIC or LION GNSS receiver
- The on-board computer implementing the station keeping manoeuvres computation and execution
The GNSS receivers addressed during the study are the MOSAIC and the LION.The MOSAIC GNSS receiver is available on the space equipment market as a qualified product and many receivers are successfully flying on satellites in low Earth orbits. It uses a software based correlator for GPS signal reception using a digital signal processor (DSP) since a hardware correlator for space application was not available on the European market at time of development. The advantage of the software based correlator is the capability of re-programming in the orbit and thus easy adaptation to changes in the modulation scheme and signal structure. The MOSAIC receiver has 8 channels.The Lion Rx is the next generation of GNSS receivers. It is designed to receive and process multi-constellation and multi-frequency signals. In particular, the Lion Rx development aims at the usage of GPS and Galileo signals (Dual Constellation capability). The Lion has a dedicated hardware for signal tracking (Hardware Correlation): the AGGA 4 (Advanced GPS and Galileo Asic 4), which will have up to 36 channels. The Lion Rx is therefore expected to track more SVs than the Mosaic Rx in general.
The station-keeping strategy is very similar to the strategy currently implemented on-ground. The required CPU-time and memory are compatible with LEON2 on-board computer unit.
The closed-loop Real-Time test bench includes the following elements:
- a MOSAIC GNSS receiver
- a LEON2 on-board computer
- a REDHAWK real-time linux PC
- a SPIRENT GNSS stimulator
Click for further details
1) State-Of-The-art in autonomous OD and SK => system level
requirements using inputs from telecom operators
2) Architecture and specification => high level requirements
3a) Design choices for orbit determination => trade-off on AOD
3b) Design choices for station keeping => trade-off on ASK
3c) Global trade-off => selection of two solutions (GNSS, APS)
4) Development of the prototype ASK SW
5) FVB and RTB development
6) FVB and RTB test campaign
7) Overall synthesis
Main requirements for an Autonomous OD/SK system were good ground ephemeris prediction based on on-board orbit measurement and on foreseen maneuvers, no overconsumption and possibility for the ground to take over by uploading orbit measured on-ground, or maneuvers plan.
Trade-off on autonomous orbit determination systems outcomes (ordered from worse to best) :
Pure orbit propagation, Sun & Earth sensor, Star & Earth, Combined Star / Earth sensor, GNSS MOSAIC, GNSS LION.
The chosen ASK strategy is the one currently implemented on-ground because it provides good visibility on the upcoming maneuvers thus allows a good ephemeris prediction, has the same fuel efficiency, is well mastered by the operators.
Validation on the Closed-Loop Functional Validation Bench (software tests) outcomes :
ASK module is validated, APS navigator is not compatible with ASK, MOSAIC and LION navigator are compatible with ASK.
Validation performed on the Closed-Loop Real-Time test Bench outcomes:
RTB includes a MOSAIC, a SPIRENT, a LEON2 implementing the ASK SW, and a RT-Linux computer, two 7 days long tests with chemical & hybrid propulsion confirmed the adequacy of MOSAIC with AODSK in nominal conditions.
Additional tests performed has shown the sensitivity of SK performances to the radial position error which could be reduced by further optimizing the GNSS receiver filter.