Geodätische Messtechnik - Prof. Dr. H. Ingensand ETH Zürichwebarchiv.ethz.ch/geometh-data/student/sensorik/... · Geodätische Messtechnik - Prof. Dr. H. Ingensand ETH Zürich Automatic

1

ETH ZürichGeodätische Messtechnik - Prof. Dr. H. Ingensand

Automatic Height DeterminationAutomatic Height Determination

On the Way to Digital On the Way to Digital LevellingLevelling

The forerunnersThe forerunnersBasics (configuration ,numerical and optical codes)Basics (configuration ,numerical and optical codes)Image processingImage processingThe individual Solutions The individual Solutions ((LeicaLeica, Topcon, , Topcon, Trimble,SokkiaTrimble,Sokkia))Performances Performances



DevelopmentsDevelopments and Patents and Patents

1966 Nivellier mit digitaler Anzeige (UNI Bonn) 1976 Sagem (Patentschrift ) 1978 Thomson (Patentschrift) 1984 Messeinrichtung zur Erfassung der Relativposition zwischen

zwei Teilen (WILD Heerbrugg). (Patentschrift) 1985 Längenmessverfahren (TU Dresden, Zeiss Jena) (Patentschrift) 1988 Anordnung zur Höhenmessung (Zeiss Jena) (Patentschrift) 1990 Wild NA2000 1990 Measuring a Difference in Elevation (Optec, Japan)

(Patentschrift)1992 Wild NA3000 1994 Zeiss DiNi10/20 und Topcon DL 100 1998 Sokkia2003 Leica DNA 03/10 Baureihe2004 Neue Trimble Baureihe auf Basis DINI (Zeiss)2004 Leica Sprinter


CategorisationCategorisation of Electronic Levelsof Electronic Levels


ActiveActive SystemsSystems

Rotationslaser Diode line


ActiveActive Systems Systems „„Das Aachener NivellierDas Aachener Nivellier““

2


Concept Study Concept Study UniBWUniBW (1984)(1984)

Chorobat


RijkswaterstaatRijkswaterstaat und DMT (WBK)und DMT (WBK)


ConceptConcept StudyStudy UNI BonnUNI Bonn

ING 1980


TheThe PriciplePriciple of of thethe „„BonnBonn““ Digital LevelDigital Level

ETH ZürichGeodätische Messtechnik - Prof. Dr. H. Ingensand ETH ZürichGeodätische Messtechnik - Prof. Dr. H. Ingensand

TheThe semisemi--automatisatedautomatisated„„BonnBonn““ Digital Level (1970)Digital Level (1970)

3


Das Bonner Das Bonner NivellierNivellier


TheThe Dresden/Jena Digital Level (1982)Dresden/Jena Digital Level (1982)


Inside the Inside the ZeissZeiss Ni002 DL Prototype Ni002 DL Prototype


TodayToday‘‘ss Digital LevelsDigital Levels


Basic Basic configurationconfiguration of DLof DL

D

H


General DesignGeneral Design

4


Image Processing inImage Processing in DigitalDigital LevelsLevels

PREPROCESSING DATA REDUCTION

FINAL IMAGE PROCESSING

OPERATING SWAPPLICATION SW

µP

OBJECT STAFF

CCD CAMERA IMAGE - A/D CONVERSION

123. 45647. 4

MAN-MACHINE-INTERFACE

A/D


The Information Transfer in DLThe Information Transfer in DL

SourceCoding

BiphasenAmplitudenPolarisation

Channel Decoding Sink

Noise

VerticalPosition

Light- artificial- natural

AtmosphereVibrations

Image processing 1.784 m


Digitales Nivellier als MesssystemDigitales Nivellier als Messsystem

Woschitz/Brunner 2001


NOISENOISE

Illumination Atmosphericinfluences

Mechanicalinfluences

Instrumentalbehaviour

Various lightintensity of naturallight (SNR)

Turbulences(blurred image,higher SNR)

Vibrations(deviation of the lineof sight)

Thermal effects(deviation of the line ofsight)

Inhomogeneous lightintensity by shadowsat the staff

Refraction(deviation of the lineof sight)

Settlement of theinstrument and staff

Interference of code-element size and pixels(wrong results atcertain distances)

Spectrum of the lightsource

Staff centring andinclination of the staff

Compensator function(eigenfrequency)


Optimisation of DL Codes PropertiesOptimisation of DL Codes Properties

1. Numerical Code Pseudo-stochastic, redundant, un-ambiguous,According to the Staff Length (0-4 m). Fast DecodingPatents of the Competitors

2.Optical Code Optimal Resolution as a Function of the Distance and the resolution of the Optics and CCD Safe against Noise Influences as: Turbulence, Blur, Image Distortions,……

Near field- , Far field CodeOptical Coding Possible(Analog or digital Width Variation, Bi-Phase,..)Patents of the Competitors


Overview of the Various Optical Codes in DLOverview of the Various Optical Codes in DL

5


Similarities of Geodetic Calculation Procedures (CodingSimilarities of Geodetic Calculation Procedures (Coding--Decoding)Decoding)

Technique Digital levels EDM GPS Biphase

Modulation X X X

Correlation X X X FFT X X


Code Code PropertiesProperties

Manufacturer Properties Near-farfieldcode

Distance/scale required

Dimension ofone codeelement

Leica Pseudostochastic Yes Yes 2.025 mmSokkia Random bidirectional

Digital width relationYes No 16 mm

Topcon Analog width variation No No 10 mmZeiss Biphase Yes No 20 mm


Image processing in DLImage processing in DL


CCD ImagesCCD Images


PixlePixle‘‘ss intensityintensity correctioncorrection byby foldingfolding withwith trapezoidaltrapezoidal functionfunction

Woschitz 2003


Analogies of Human and Technical Image Processing Analogies of Human and Technical Image Processing

Lateral Inhibition Radiometric Lateral Inhibition Radiometric CorrectionCorrection

Edge Operators

Canny

Sobel

...........

6


AncillaryAncillary DevicesDevices: : InvarstaffsInvarstaffs


InvarstaffInvarstaff GeometryGeometry

Position of Invar BandCenter of Baseplate

Body(Aluminium)

Invar Band

10mm


CenteringCentering Rings Rings fromfrom NEDONEDO


Leica Leica FamilyFamily NA 2000/3000 DNA03/10NA 2000/3000 DNA03/10


TheThe newnew Leica Digital Level Leica Digital Level FamilyFamily


The WILD The WILD LeicaLeica Development NA 2000/3000Development NA 2000/3000

7


SectionSection of of thethe NA2000NA2000


InsideInside thethe NA2000NA2000


Design Design –– Inside Inside LeicaLeica LevelLevel

Magnetic dampened compensatorThe use of incandescent or halogen lighting is also possible.


TheThe Leica CodeLeica Code

Thesis Woschitz


TheThe CorrelationCorrelation ProcedureProcedure of Leicaof Leica


TheThe ScaleScale ProblemProblem

8


TheThe CorrelationCorrelation „„MountainMountain““ of of thethe Leica DL Leica DL

Two dimensional correlation

Cor

rela

tion

Coe

ffici

ent

Height

Distance/ Scale

signalreferenceHDPsignalQ

HDPQn

HD iiQP

),(

),(1),(256

1

==

•= ∑ρ


The Determination of the Best CorrelationThe Determination of the Best Correlation

Distance

Height

0.000 m

4.050 m

About 50 000 correlation in the whole range

0.00 m 100.00 mFocus Position

One bit correlation


TheThe One Bit One Bit CoarseCoarse CorrelationCorrelation

operatorEXNORsignalreferencehdP

signalQ

hdPyQn

hd i

n

iQP

),(

),()(1),(1

1

=⊗==

⊗= ∑−

ρ


Normalized 8Normalized 8--bit correlationbit correlation

21

1

221

1

2

1

1

11

1

),(PP

nQQ

n

PQPQnHD

n

i

n

i

i

n

i

QP

−•−

•−•=

∑∑

∑−−

−

ρ

signalreferencehdPsignalQ

),(

==


Technical Data DNA03/10Technical Data DNA03/10

Standard deviation per km double levellingWith invar staff: 0.3mm / 0.9mm

Field of ViewCoarse correlation – 2° at 50mFine correlation – 1.1°

RangeStaff: 0 – 4.05mDistance: 1.8m – 110m

Standard deviation of distance measurement5mm/10m (500ppm)


9


ZeissZeiss/Trimble /Trimble DiNiDiNi


InsideInside thethe Zeiss Zeiss DiNiDiNi


OpticalOptical Layout of Layout of thethe Zeiss/TrimbleZeiss/Trimble


The The ZeissZeiss CodeCode

a 30 cm section of the rod gives a 1 byte information

Coarse code

Near field code


TheThe Zeiss Zeiss BiphaseBiphase CodeCode


Zeiss CodeZeiss Code

Thesis Woschitz

10


SamplingSampling of of thethe CodeCode

Woschitz/2003


GeometricGeometric MethodMethod ((ZeissZeiss))

STAFF IMAGE

DIGITISING


Determination of the Scale of the Image (Determination of the Scale of the Image (ZeissZeiss))

Df

GB

D B f Gi

ii i= ⇒ ⋅ = ⋅

Distance1515

=>−⋅

=−+ BB

gNA

Using a +/-15 cm Interval:

According the Reichenbach method with stadia lines


The The ZeissZeiss Calculation ProcedureCalculation Procedure

Gi

Gi+n

Bi

]2

)2/1([1 111

0

bbACgN

H iN

ii

+⋅−+⋅= +

−

=∑

Number of Codeword (Lower edge)Position of a

single interval

Position of the optical axis in the image (CCD)

Middle of the code bar

gi


Problems Problems withwith thethe Zeiss/Trimble Instruments Zeiss/Trimble Instruments


Two Epoch HeightTwo Epoch Height--Differences versus HeightDifferences versus HeightDifferences 99-98

Height [m]-800-700-600-500-400-300-200-100

0100

2 10 18 26 34 42 50 58 66 74 78 82 90

2959

K30

96K

3241

K10

611

412

2

130

138

4605

K14

6

Positions

[1/1

00 m

m]

730740750760770780790800810820

11


Experimental Setup (ETHZ 2000)Experimental Setup (ETHZ 2000)


Intensity Distribution Intensity Distribution

Spot Illumination

0 50 100 150 200 2501

276

551

826

1101

1376

1651

Pix

el

Intensity


Inhomogeneous Illumination at BackInhomogeneous Illumination at Back-- and Foresight Shotand Foresight Shot

SpotSpot


The levelling staffThe levelling staff--end problem superimposed with end problem superimposed with inhomogeniousinhomogenious illuminationillumination

Deviations as a function of the position and illumination: Landestopographie top

down Software version.: 1.05

1.601.701.801.902.00

-0.20 0.00 0.20 0.40[mm]

[m]

Deviations as a function of the position and illumination: Landestopographie down-top

Software version.: 1.05

1.60

1.70

1.80

1.90

2.00

-0.01 0.04 0.09 0.14[mm]

[m]


New New StaffStaff Illumination Illumination byby Riesen& Riesen& StettlerStettler

Stettler&Riesen

Stettler&Riesen

Solothurn


Temperature Function of Temperature Function of ZeissZeiss DiNiDiNi ( UNI Karlsruhe)( UNI Karlsruhe)

12


TheThe TopconTopcon PrecisionPrecision Digital LevelsDigital Levels


The Topcon Digital LevelThe Topcon Digital Level


TheThe TopconTopcon DL103 (Construction Level)DL103 (Construction Level)DL-103 Specifications

Telescope:26x30mm 1°30'4 0.9m Erect Image

Compensator:±10’ 0.5“

HeigtAccuracy (Standard Deviation for 1km double run leveling) ±1.8mm w/Fiberglass Staff ±2.0mm w/Aluminum Staff±2.5mm0.1mm/1mm (0.001ft/0.01ft) selectable

Distance Measurement:1cm(0.02ft)

Accuracy (D= measuring distance: m) ±(0.1% x Dm) (D>10m),±10mm (D<10m)±(0.15% x Dm) (D>10m),±15mm (D<10m)2m to 60m (7ft to 197ft)2 sec.10'/2mm

Others:128 x 32 dot, Dot matrix LCD with backlight (55.01mm x 16.29mm)

RS-232C5

Rod height/difference of elevation

Built-in timer for auto power–off360° or 400gonAA battery x 420 hours(Alkaline)-20°C to +50°C (-4°F to +122°F)IPX 6196(H) x 158(W) x 232(L) mm7.7(H) x 6.2(W) x 9.1(L) inch2.3kg

StaffLength: 3m[9.84ft] Length: 5m [16.40ft] Length: 5m [16.40ft]


The Topcon CodeThe Topcon Code


TheThe SOKKIA SDL30SOKKIA SDL30


TheThe SokkiaSokkia CodeCode

Close Range

5 Value Resolution

Long Range

3 Value Resolution

13


Pattern width(mm)

0-5 code(Short dist.)

0-2 code(Long dist.)

Image size at 100m

Image size at 10m

3 0 0 1.05µm/ 1 cell 84µm/ 10.5 cell

4 1 0 1.40µm/ 1 cell 112µm/ 14.0 cell

7 2 1 2.45µm/ 3 cell 196µm/ 24.5 cell

8 3 1 2.80µm/ 3 cell 224µm/ 28.0 cell

11 4 2 3.85µm/ 4 cell 308µm/ 38.5 cell

12 5 2 4.20µm/ 4 cell 336µm/ 42.0 cell

Relations between pattern width, code and distance.

Properties of the Bar Code of SDL30


Overview of the Overview of the PerformacesPerformaces of DLof DLInstrument TOPCON LEICA SOKKIA ZEISS/TrimbleFeature DL102 NA3003/DAN02 SDL30 DiNi10/11/12

Accuracy mm/Km Double levelling 0.4 mmInvarstaff

0.4 mmInvarstaff

1.0 mmFibreglassstaff

0.3 mmInvarstaff

Distance (Resolution) 1 cm 1 cm 0.1% x D 1 cm

-Pendulum

0.3"± 15'

Pendulum0.3"± 15'

Pendulum-

> ± 15'

Pendulum0.2"± 15'

Measurement time 4 s 4 s > 3 s 4 s

Range 2-60 mInvarstaff

1.5 - 60 mInvarstaff

1.6 – 100 mStandard-staff

1.5- 100 mInvarstaff

Man- machine-interface Menue Menue with function keys

Menu Menu withfunction keys

Display 2 lines 2 lines 2 lines 4 lines

Operation time/ battery 10 hours 8 hours > 7 hours 1 day

Weight including battery 2.8 kg 2.5 kg 2.4 kg 3.0 kg

Required Field of View noinformation

2° 1°20' Minimum 30 cm

Data storage capacity 2400 500/ Depends on Memory Card

- 2000

Compensator TypeAccuracyRange


ApplicationsApplications


Permanent Monitoring with Permanent Monitoring with MotorisedMotorised Digital LevelsDigital Levels


AircraftAircraft LevellingLevelling


CalibrationCalibration of Digital Levelsof Digital Levels

14


Digitale Nivelliersysteme: SystemkalibrierungDigitale Nivelliersysteme: Systemkalibrierung• Präzision der TUG Anlage ist besser als 3 µm.

• Beinhaltet auch den nichtlinearen Zusammenhangder Komponenten (z.B. Lattenzustand in Kombination mit Auswertemethode).

Nur die Systemkalibrierung bestimmt die Gesamtpräzision.


Digitales Nivelliersystem: Schema der SystemkalibrierungDigitales Nivelliersystem: Schema der Systemkalibrierung


Ansichten der TUG AnlageAnsichten der TUG Anlage


Software zur Systemkalibrierung (1)Software zur Systemkalibrierung (1)


Software zur Systemkalibrierung (2)Software zur Systemkalibrierung (2)


Ablauf der Systemkalibrierung fAblauf der Systemkalibrierung füür digitale Nivelliersystemer digitale Nivelliersysteme

• Abtastintervall

I ... auf die Latte projizierte Länge des CCD bei 3 m

• Hinmessung

• Rückmessung um I / 6 versetzt

TA = I / 3

(Leica: I = 117 mm, Zeiss: I = 262 mm)

15


LinearitLinearitäätsabweichungen: 2 Lattentsabweichungen: 2 Latten• Leica NA3003, TA = 39 mm• Entfernung: 3 m• STD = ±14 µm


LinearitLinearitäätsabweichungen: 2 Tagetsabweichungen: 2 Tage• Leica NA3003, TA = 39 mm• Entfernung: 3 m• Latte A• STD = ±14 µm


Resultate der SystemkalibrierungResultate der Systemkalibrierung

• Reproduzierbare Nichtlinearitätsabweichung bei 3 m:- Leica: unabhängig von Latte und Messungen !- Nivellier ?, Code ?


Resultate der SystemkalibrierungResultate der Systemkalibrierung

• Reproduzierbare Nichtlinearitätsabweichung bei 3 m:- Leica: unabhängig von Latte und Messungen !- Nivellier ?, Code ?

Documents

Geodätische Messtechnik - Prof. Dr. H. Ingensand ETH Zürichwebarchiv.ethz.ch/geometh-data/student/sensorik/... · Geodätische Messtechnik - Prof. Dr. H. Ingensand ETH Zürich Automatic