The new ISO standard 1713-8 for checking GNSS field measuring systems Hans Heister Institute for Geodesy - Geodetic Metrology - UniBw München D-85577 Neubiberg h.heister@unibw.de FIG Working Week - Stockholm - Sweden, 17 June 008 1 Outline General remarks Checking and calibrating of GNSS systems The ISO series of standards 1713 The new standard ISO 1713 part 8 Additional proposals for checking GNSS systems Conclusions 1
General Remarks Checking and testing of geodetic instruments: Four phases model (FIG): Phase 1 Simple functional check: Objectives / Operations Evaluation of operability, visual inspection, short intervals before and after measuring, in the field. GNSS Phase Etended functional check: Simple quantitative checking of significant deviations of specified thresholds in regular intervals or event dependant. () 3 General Remarks Checking and testing of geodetic instruments: Four phase model (FIG): Phase 3 Calibration: Objectives / Operations Nominal /actual value comparison of the defined measurands.. Measuring reference (standard), special calibration facilities, traceability, certificate, fied intervals, epenses GPS? Phase 4 Specification test, type testing: Checking of the technical specifications in? conjunction with QM, independent testing of new instruments or prototypes, components? and / or measuring systems, knowledge of measuring principles and software /? firmware, event dependant, manufacturers certificate. 4
General Remarks What makes checking of GNSS measuring systems difficult? black bo system! soo Quelle : Fuhlbrügge 5 General Remarks Original measuring quantities of GNSS measuring systems? 3D coordinates X Y Z in WGS 84 Product in the sense of a QM (ISO 9000) Accuracy / measuring uncertainty Correctness Reliability System integrity 6 3
Checking and Calibrating of GNSS Systems Measuring procedures RTK Local area, Local, temporary reference station Post processing Influence quantities Wide area, Nationwide, permanent reference station, networked Kinematic applications, e.g. machine guidance, mobile mapping Influence quantities System Calibration / Calibration / Checks of Checking instrumental components 7 New standard ISO 17 13-8 GNSS field measurement systems in real-time kinematic (RTK) Concept of the test procedures a R R 1 b m < a < 0m b according to the task B Horizontal distances and height differences determined by tacheometry, s d < 3 mm Nominal values D*, h* Observables, measurements: Coordinates, y ellipsoidal height h (WGS 84) D, h Comparison D* D* D and h* h* h h 9 4
New standard ISO 17 13-8 Procedure 1: Simplified test procedure R 1 a b R m < a < 0m b according to the task B Limited number of measurements 1 Series 5 Sets Measurements at R1 and R 1,1,1 1,1, 1,,1 1,, Time lag 5 min! D ( ) + (y y ) i,j i,j, i,j,1 i,j, i,j,1 h h h i,j i,j,1 i,j, 1,5,1 1,5, ~ 5 min ( y h ) i,j,k i,j,k i,j,k i,j,k 10 New standard ISO 17 13-8 Procedure 1: Simplified test procedure ε D Di,j ε h hi,j i,j i,j D * h* ε Deviations of horizontal distances and height differences Standard Deviation s,y a) Specified by manufacturer b) determined by full test procedure ε ε Di,j hi,j,5 s,5 s y h If If any deviation fails the the two two conditions, the the inclusion of of outliers is is suspected!! Repeat test procedure!! 11 5
New standard ISO 17 13-8 Procedure 1: Simplified test procedure Objective: Determination of the operational reliability of the GNSS-equipment and a simple quantitative check carried out under minimal eterior influences and minimal effort (FIG phase 1) Procedure : Full test procedure Objective: Etended functional check, quantitative investigations of significant deviations of specified thresholds (statistical evaluation) in regular intervals or event dependant (FIG phase ). 1 New standard ISO 17 13-8 Procedure : Full test procedure Objective: Best achievable measure of of precision eperimental standard deviation for a single position, height Higher measuring effort 3 Series 5 Sets Measurements at R1 and R 1,1,1 1,1, 1,,1 1,,,1,1,1,,,1,, 3,1,1 3,1, 3,,1 3,, 1,5,1 1,5,,5,1,5, 3,5,1 3,5, Time lag > 90 min! ~ > 3 h 151 5 min ( y h ) i,j,k i,j,k i,j,k i,j,k 13 6
New standard ISO 17 13-8 1. Step: Individual measurements are compared with the nominal values. Same evaluation as described in the simplified test procedure Objective: Detection of gross errors. Step: Statistical evaluation Estimates, y, h of R1 and R 1 k 15 3 5 i1 j1 i,j,k y k,h k r i,j,k k - i,j,k r y i,j,k,rh i,j,k Residuals for all measurements Degree of freedom ν ν y ν h (3 5 1) 8 14 New standard ISO 17 13-8. Step: Statistical evaluation Standard deviation of a single measurement s r ν r 8 s y r y 8 s h r h 8 Eperimental standard deviation for a single position (,y) Eperimental standard deviation for a single height (h) s ISO-GNSS RTK y s + s y s ISO-GNSS RTK h s h 15 7
New standard ISO 17 13-8 Step 3: Statistical tests Is the calculated s ISO-GNSS RTK y smaller or equal than the value y stated by the manufacturer? σ y s ISO-GNSS RTK y σ y χ ( ν +ν ) 0,95 X y ν +ν X y χ (56) 74, 47 0,95 s ISO-GNSS RTK y y σ y 1,15 Is the calculated s ISO-GNSS RTK h smaller or equal than the value σ h stated by the manufacturer? s ISO-GNSS RTK h σ h 1, 17 New standard ISO 17 13-8 Step 3: Statistical tests In the case of two samples, the Fisher test indicates whether two eperimental standard deviations e.g. s ISO-GNSS RTK y and s ISO belong to the same population? ISO-GNSS RTK y F 1 α 1 ( ν % +ν % y+ν +νy) s s% ISO GNSS RTK y F α 1 ( ν % +ν % y +ν +νy ) ISO GNSS RTK y F 0,975 (56,56)1,70 s 0,59 1,70 s% ISO GNSS RTK y ISO GNSS RTK y s 0,47,13 s% ISO GNSS RTK h ISO GNSS RTK h 18 8
New standard ISO 17 13-8 Questions to be discussed Time need of of 3 4 hours reasonable? Number of of series of of measurements sufficient? Fiing of of allowable thresholds? Disadvantages of of the the procedures? Appliance of of the the tests tests on on RTK RTK using reference station networks possible? Is Is the the indication of of the the eperimental standard deviation strong enough for for characterizing quality of of the the investigated measuring equipment? 0 Conclusions The GNSS measuring system as a non autonomous and black- bo system is for the practitioners hard to understand. Periodical checks or tests for GNSS systems in terms of a quality management system are unavoidable. In this sense applicable, standardized field checks RTK -, not calibration, are urgently needed (FIG phase 1 and ). The new ISO standard 1713-8 8 is a first move. Antenna checks / calibration demand further discussions (specifications, manufacturers information). recent New methods, techniques (kinematic( kinematic!) und instrumental developments e.g. GPS-tacheometer, frequencies, etc. permit improved / etended test approaches. 6 9