NRTK services in Ireland - an Evaluation

Transcription

1 NRTK services in Ireland - an Evaluation Dr. Audrey Martin & Dr. Eugene McGovern, Dublin Institute of Technology, Ireland. FIG Working Week, May 12 Ireland s Survey Infrastructure 1995 IRENET ING 185 Ground Points ± mm 2 IRENET2 ITM CORS RINEX Output ± mm 4 Active CORS Network NRTK 12 Stations GPS only GSM 1

2 Ireland s Survey Infrastructure 6 Active CORS Network GPRS 9 Active CORS Network 16 Stations GPS + GLONASS Stations Planned: OS IE/UK9 OSGM9 ~8 Infill stations NRTK Users NRTK 2

3 Advantages of NTRK No base station(s) required Users only require one receiver Increases productivity Reduces outlay High quality Infrastructure Geodetic quality GNSS receivers Calibrated choke ring antennas reduce multipath Direct access to National Coordinate Reference Frame Integrity monitoring NRTK Service Providers in Ireland SIS (for OSi) 23 stations MAC solution Korec Irl. 22 stations VRS solution Topcon Irl. Smartnet stns MRS solution 3

4 MAC Solution (Master Auxiliary Concept) Data hub receives streaming data from CORS network Fixes ambiguities between master station and auxiliary network stations Estimates ionospheric, tropospheric and orbit error effect at each station Sends to rover full observation and coordinate information for a master reference station plus data for all auxiliary stations Rover generates approximate coordinates Interpolates ionospheric, tropospheric and orbit errors for location Ambiguities fixed between rover and all network sites and position determined Significant computational burden at rover VRS & MRS Solutions (Virtual/Modelled Reference Station) Data hub receives streaming data from particular CORS network Rover computes approximate position and transmits it to the data hub Data hub then - models the ionosphere, troposphere & orbit effects for the rover location - generates a data stream for the specific rover location These data then becomes the virtual reference station data VRS data transmitted to the rover VRS site becomes the fixed base station very short baseline to rover, therefore ambiguity resolution is easy Reduced requirement for sophisticated software in rover Traceability may present a QA issue 4

5 Research Objective Get a snapshot of the performance of the three NRTK systems under typical hire-out conditions. i.e. Standard hire-fleet equipment Used as-is No mission planning No further procesing Leica NetRover Trimble R8 Topcon GRS-1 Methodology 9 IRENET points Simultaneous recording Epoch rate = 5 secs Cut-off angle = Measurement time = ~45 mins GLONASS then No further processing Results compared to IRENET coords 5

6 Results - 2D 2D RMSE - Glonass included 5 mm 3 Leica Trimble Topcon IRENET Station 2D RMSE - Glonass not included 5 mm 3 Leica Trimble Topcon IRENET Station Results - 2D GLONASS / Leica - Glonass on/off 2D RMSE (mm) 5 3 Trimble - Glonass / 2D RMSE (mm) 5 3 2D RMSE (mm) 5 3 Topcon - Glonass / 6

7 Results - Height Height RMSE - GLONASS included RMSE (mm) Leica Trimble Topcon Topcon (Oct. 11) IRENET Station Height RMSE - GLONASS not included RMSE (mm) Leica Trimble Topcon Topcon (Oct. 11) IRENET Station Results - Height GLONASS / Leica - Glonass on/off Height RMSE (mm) D137D138D145D146D147D151D152D156D158 Height RMSE (mm) Trimble - Glonass on/off Height RMSE (mm) Topcon - Glonass on/off D137D138D145D146D147D151D152D156D158 7

8 Results - Initialization Issues 2,5 2 Residual (m) 1,5 1,5 de dn dht : 1: 2: 3: 4: 5: 6: 7: 8: 9: : 11: 12: 13: 14: 15: 16: 17: 18: 19: : 21: 22: 23: 24: 25: 26: 27: 28: 29: 3: 31: 32: 33: 34: 35: 36: 37: 38: 39: : 41: 42: 43: 44: Time (mm:ss) Example of a loss of initialization and failure to correctly reinitialize at D138. Results - Windowing,6,, (m), -, -, : 1:15 2:3 3:45 5: 6:15 7:3 8:45 : 11:15 12:3 13:45 15: 16:15 17:3 18:45 : 21:15 22:3 23:45 25: 26:15 27:3 28:45 3: 31:15 32:3 33:45 35: 36:15 37:3 38:45 : 41:15 42:3 43:45 45: 46:15 47:3 48:45 E N Ht -,6 -,8 -, Example of variability of residuals with time for one observing session 8

9 Results - Averaging,15,,5 (m), -,5 -, : 1:15 2:3 3:45 5: 6:15 7:3 8:45 : 11:15 12:3 13:45 15: 16:15 17:3 18:45 : 21:15 22:3 23:45 25: 26:15 27:3 28:45 3: 31:15 32:3 33:45 35: 36:15 37:3 38:45 : 41:15 42:3 43:45 45: 46:15 47:3 48:45 E N Ht -,15 -, -,25 Example of accumulated mean residuals for one observing session. Results - Elevation Difference Results for D156 were comparable to other locations. IRENET Pt. D156 in Wicklow Mountains (El. 488 m) A significant initialization issue did, however, occur. 9

10 Conclusions With respect to measurements taken at the nine IRENET stations. Three NRTK systems deliver comparable accuracies The addition of GLONASS did not make a difference Initialisation issues can occur Averaging and Windowing may not improve results Sig. elevation difference was not an issue XY accuracy = 22 mm ±8 mm Z accuracy = 29 mm ±14 mm Acknowledgements The authors wish to thank the following for their support: Society of Chartered Surveyors Ireland Survey Instrument Services (Leica) Korec Ireland (Trimble) Topcon Ireland (Topcon)

11 Thank you for your attention 11