GAST-D Monitoring Results from Post- Processed Flight Trial Data - Performance Evaluation of DLR's GBAS Testbed

摘要

The German Aerospace Center (DLR) implementedan experimental GBAS test bed at Braunschweig-Wolfsburg Research Airport. Its compliancy withrespect to accuracy, integrity, and availabilityrequirements of CAT-I precision approach has beendemonstrated in several flight trails in 2009.The research ground station consists of threeTopcon Net G3 dual frequency GNSS receiverswith multipath suppressing choke ring antennas,which deliver correlation results for navigation at arate of up to 20 Hz. However, in the current stageonly the certified L1 range and carrier data is used.The high sampling rate was chosen in order toeffectively detect cycle slips in the phase data.Stations are separated by baselines between 740and 770 meters. Due to its high modularity, thestation is fully accessible and can be easilyreconfigured for research purposes. Currently,reconfiguration to an enhanced infrastructure istaking place to comply with the more stringentrequirements of CAT-III / GAST-D approaches.In preparation for this reconfiguration, we postprocessedthe data collected during the 30approaches of the 2009 flight trials (flown withDLR’s VFW614 research aircraft) with respect tothe GAST-D parameters and requirements definedin the ICAO draft SARPs and RTCA DO-253C.These requirements include a second smoothingfilter and pseudorange correction processing with30s smoothing time on both, ground and aircraftsubsystem as well as a battery of new monitors onthe airborne side. These monitors include the dualsolution ionospheric monitoring architecture,differential correction magnitude check, biasapproach monitor, reference receiver fault monitorand fault detection and exclusion. We show theoutputs of these monitors during the flight trials andevaluate their performance with the one expectedtheoretically. We overbound the output of theDSIGMA monitor in order to obtain inflatedparameters σ_V=0.24 m and σ_L=0.17 m for thereference receiver fault monitor.We evaluated integrity, availability and continuityperformance during the flight trials. For this weused dual frequency carrier phase positioning in acombined forward and reverse solution as a truthreference. At no time misleading information orhazardously misleading information occurred andthe service was never interrupted or unavailableduring any of the approaches. Position accuracywas below 1,86 m vertically and 0.91 m laterally atall times within the GBAS service area.Moreover, we have optimized the location ofground station antennae to fulfill the ICAOrequirement of detecting absolute ionosphericgradients above 300 mm/km with the monitoringarchitectures under the assumption of a worst caseplacement of the reference station 5 km away fromthe threshold as suggested by Khanafseh et al.(2010). The optimized configuration for fourreceivers is linearly distributed along the runwaywith a maximum baseline of 220 m and allows anincrease of allowable carrier phase noise by 20%and therefore to a standard deviation of 8 mm.