Geodetic Challenges on Precise Positioning & Navigation ... · GNSS Raw-Data on Smartphone => New...

Geodetic Challenges Reiner Jäger GSA Task Force GNSS-Raw-Data on Smartphones Prague, 30-May-2018

Geodetic Challenges on Precise Positioning

& Navigation using GNSS Raw Data on

Smartphones along the NAVKA-Disy RaD

Project PREGON-X

Hochschule Karlsruhe Technik und Wirtschaft (HSKA) - University Applied Sciences

Laboratory on GNSS & Navigation

http://goca.info/Labor.GNSS.und.Navigation/index.php

RaD

www.dfhbf.de, www.goca.info, www.geozilla.de, www.navka.de

Email: [email protected]

Reiner Jäger

http://goca.info/Labor.GNSS.und.Navigation/index.php

http://www.dfhbf.de/

http://www.goca.info/

http://www.geozilla.de/

http://www.navka.de/


Introduction

NAVKA (Navigation Technologies Karlsruhe) - Projects

GNSS/MEMS/Optical based State Estimation

Sensors providing Space / Parameter Relation

B.W. ZIM Network High Precise Realtime Navigation


GNSS / MEMS / MOEMS Algorithms and Software for Out-/Indoor Navigation (People, Vehicles, Goods) and Georeferencing with

distributed Sensors and Platforms

MOEMS

RaD Project

www.navka.de

https://www.youtube.com/watch?v=-k--3GxrQXU

Tbzeb

byeb

bxeb

eeeeeeeeeeee z y xy p r zyxz y xy s||||||,,,


https://www.youtube.com/watch?v=-k--3GxrQXU


NAVKA-Projects

www.navka.de



NAVKA-Projects www.navka.de www.volocopter.com



Since April 2018 ZIM-Network High Precise Navigation B.W.

http://navka.de/index.php/en/weit

ere-projekte-3/project-zim-1de

http://navka.de/index.php/en/weitere-projekte-3/project-zim-1de


NAVKA – Sensor - Leverarm Consideration


Accelerometer- Sensorobservation j (Rawdata, on i-th Platform (p) – one-dimensional!)

ipbp ],[ Rt

i

j

j,iss )],(,[ rt

Orientation j-th

Sensor (s)

on i-th Platform (p)

j,i

pij,si

sin

sincos

coscos

r

j,i

i

ijj,i

see

ieeie

eeie

ee

be

i,pb

p

ss

])t( )t(2 )()t([

)y,p,r(a

xΩΩxΩxgx

RRr

i

j,i

i

j,ij,i

p

s

p

ssa ar i,es

be

i,pb

p

s j,i

i

j,i)y,p,r( aRRa

Observation-Equation for Sensor j

on Platform i

Sensors GNSS/MEMS/OPTICAL - Multisensor/Multiplatform


][)y,p,r(j,i

ps,p

bp

bp,b

eb

eb

es

i

iiij,itRtRxx

j,i

i

ijj,i

see

ieeie

eeie

ee

be

i,pb

p

ss

])t( )t(2 )()t([

)y,p,r(a

xΩΩxΩxgx

RRr

Referencing : Platform p(i) on Body (b) and Sensor s(i,j) on Platform (i) – „Leverarms“

(1)

(2)

„l(i,j)“

Observation-Equation for Sensor j

on Platform i

Sensors GNSS/MEMS/OPTICAL - Multisensor/Multiplatform

bz,eb

by,eb

bx,eb

bz,eb

by,eb

bx,eb

eeeeeeeeeeee | |z y x||y p r| zyx|z y x)t( y

Tbz,eb

by,eb

bx,eb

eeeeeeeeeeee |z y x||y p r| zyx|z y x)t( y

From 15 to 18

Parameters y(t)

Change in StateTransition

Model and Estimation on y(t)


NAVKA – GNSS and Smartphone Related Developments


GPS Orbit and GlONASS Orbit

Chinese

COMPASS Orbit

Orbit – Segment Galileo

28 Dec.2005 Giove-A lift off

05:19 UTC Baikonour,

Kazakhstan Galileo Satellite

Smartphone and Tablets for Mobile GIS and Geo-Referencing

GNSS Raw-Data on Smartphone => New Age, in 2017/2018


OPPP – E1/E5 and

L1/L5

Phasemeasurements

– 21. Sept. 2017 -

1 cm

OPP-K global

Positioning

• GNSS-Receivers

• Smartphones

GNSS receiver

„BCM47755“


GNSS Raw-Data on Smartphone => New Age, in 2017/ 2018


1.) MEMS-Gyro

Observation l:

nin

pn

pip

pnp )y,p,r( ωRωω

pipω

2.) MEMS-Accelormeters

Observation l: ap

pip

ii )t()()t( aRxgx

New Way SatNav&MIT B.W. Joint RaD Project (NAVKA):

Autonomous MEMS-Sensors + New Algorithms „Deep-Coupling“

2. Autonomous

References: Inertial Space (i) and Gravity Field


4.) Magnetic SensorsAuxiliary Sensor

Reference: Earthmagnetic Field MReference: Earth Atmosphere

2. Autonomous


GNSS Raw-Data on Smatphone – Variants 1.2 and 3


Camera

Global System Xe, Camera (c) System X‘ and Point Pk

)(),,(

'z

'y

'xO,ek,eTe

c

k

xxR

Central-Projection Camera-System

k

k

k,c

'y

'x

'z

c

cv

cu

y

x

k

kk,c

'z

'y

'x

c

cv

cu

y

x


NAVKASMART-

Phone Marker

APP (Optical

Markers, LED,

Infrared)


GNSS Raw-Data on Smatphone – Variants 1.2 and 3

c

c


)(tVOv

bVOt

Final Observation Equation in Tight Coupling

1.)

Velocity - Observation from the differentation of

subsequent VO-based position-differences

General leverarm-situation for Camera

VObzyx rrr ),,(R

)t(ωRxRRvb

eb

e

b

e )y)p,(r,(t)(y)p,(r,(t) b

VO

b

e

VO

b

VO


GNSS Raw-Data on Smartphone – Variants 1.2 and 3


2.) Orientation-rate observation from the subsequent VO-based orientation change

General leverarm-situation for Stereo-Camera

)(, tVOVOeΩ

Final Observation Equation in Tight Coupling

bVOt

VObzyx rrr ),,(R

TVOb

0

bVOb,

bbe,

0

beFrame,Object

VOb

TVOb

bVOFrame,Object

VOb

VOVO,Frame"Object"

)()(t)(t)(t)(

)((t))((t)

RΩΩΩR

RΩRΩ


GNSS Raw-Data on Smartphone – Variants 1.2 and 3


NAVKA – State Estimation


Tbz,eb

by,eb

bx,eb

eeeeeeeeeeee |z y x||y p r| zyx|z y x)t( y

....]t)(!2

1tI[)tt()t( 22b

ebbeb

be

be ΩΩRR

Orientation / Attitude

bie

bib

beb )Sensor( ΩΩΩ with

Space Curve of the body (b ) in the e-frame

)tt()t( beb

beb ΩΩ

Rotation Rates of the Body (b ) with respect to the e-frame

State Transition-Equations for the body (b) in the e-frame

Modifcation

of the State

Parameters

and

Equations

by

Considering

Special Conditions

Detected in

Multithreading

Computing

NAVKA Multisensor-Multiplatform Leverarm Design


NAVKA Multisensor-Multiplatform Leverarm Design


0yF )t(( oyFu ))t((

Zero-Updates („ZUPT“) „Indoor-Map-Matching“„Automotive Mode“

Roll

Velocity

General M-Estimation & Additional State InformationRobust Kalmanfiltering using L1-Norm Simplex Algorithms

Parallel Processing Algorithms

Condition Equations Condition In-Equations

3.3) Sensorintegration

Tachometer MOEMS

Laser-

Scanner

SLAM

Markov-Chain

& Bayes -

Approach


ZIM Cooperation Project on Mobile GIS

PREGON-X

- Precision GNSS and Georeferencing on Smartphones, Tablets etc. = x –

-

Funded as ZIM (Zentrales Innovationsprogramm Mittelstand) project by the

German Ministry on Economy and Energy ((Bundesministerium für Wirt-

schaft und Energie).

NAVKA-Project IAF/HSKA

www.navka.de

and DISY

(www.disy.net)

PREGON-X


http://www.disy.net/


ZIM Cooperation Project NAVKA DISY

on “Mobile GIS”

Precise GNSS/Georeferencing On Smartphones

(PREGON-X)


PREGON-X Smartphone for Mobile GIS and Geo-Referencing

Variants 1.1, 1.2, 2 and 3


2

3

4

TRIMBLE

VRSNOW

4 RTCM Services: 1-3 cm Horizontal and Height

Geodetic Infrastructures for GNSS-Services (GIPS):

www.moldpos.eu

www.geozilla.de

1

SRPOS Siberia


GNSS Raw-Data on Smartphone => From DGNSS to PPP-K

http://www.moldpos.eu/

http://www.geozilla.de/


International GNSS-Service (IGS „RTS“: http://rts.igs.org/




August 2017: Joint Venture „Sapcorda Services“ (Bosch, Geo++, Mitsubishi, ublox)





Survey on Existing Solutions

Limitation concerning Direct Object Access



Variants 1.2 and 3

bz,eb

by,eb

bx,eb

bz,eb

by,eb

bx,eb

eeeeeeeeeeee | |z y x||y p r| zyx|z y x)t( y



Survey on existing Solutions. 2.) Contact-free Georeferencing

Not new (using laser-distance unit or optics/camera !)

http://www.ikegps.com/spike/

With lase already „done

in 2012“ by NAVKA

http://www.navka.de/ind

ex.php/en/overview

• PPP-K not operable

• In „advance to time“

Current

Master-

Thesis

http://www.ikegps.com/spike/

http://www.navka.de/index.php/en/overview


NAVKA

Hardware Developments

in

PREGON-X


As the diagram denotes,the data from the GPS receiver ublox M8T is transmitted to the Bluetooth module NINA -B112. The power supply can be provided through the USB port or through the backup power supply through the battery on board.

PREGON-X

Smartphone

for Mobile GIS

and Geo-

Referencing

Variant 1.1


PREGON-X Smartphone for Mobile GIS and

Geo-Referencing - Variant 1.1 – EAGLE Layout


PREGON-X Smartphone for Mobile GIS and Geo-

Referencing - Variant 1.1 – GNSS/BT Board


GNSS and Challenges


GNSS Code- or „Pseudorange“ Observation and Modeling

Pseudorange or Code-Observation

))()(())()(()( ,, jSjiRijjSi

iRObs

jiObsi

ji ttttcttttcct



Pseudorange Modeling in ECEF and GNSS-time

))()(())()(()( ,, jSjiRijjSi

iRObs

jiObsi

ji ttttcttttcct

)()( jjSi

iRObs

ji tttt

jSiRjijjSi

iRObsi tctcttcttttct ,,)())()(()(

jSiRj

jiObsi tctcttt ,,),,(~)( o

jSiRj

jjSi

IRObsi tctcttt ,,),(

~)(

~)( oxx

jSiRj

iijSi

IRObsi tctcttt ,,),(

~)(

~)( oxx

jSiRji

jí

ijSi

IRObsi tctc

c

tttt ,,),

)(~(

~)(

~)(

oxx

with

))(~

(~

)(~

)(~

c

tttt i

jí

iSiRi

xxwith

1. Without Atmosphere corrections



2. With Atmosphere (Ionosphere and Troposphere)

jTROPi

jIONijSiR

jijí

iSiRObsi tctcc

tttt ,,,,),

)(~(

~)(

~)(

oxx

))(~

(~

)(~

)(~

c

tttt i

jí

iSiRi

xxwith



GNSS Positioning and Modeling

3. Geodynamic Corrections

1. 3D-Earth Tide Corrections

3.2. Ocean Loading (IERS Standards, 1996)

3.3. Atmospheric Loading

3.4. Earth Orientation

Pr)( RRRRR NEPei t )t()t()t( ie

ie

xRx

Equation" Time Light" :Iteration - t

jijí

iSiR

ijí

c

ttt

)(~

)),)(~

(~

)(~

oxx

Light Time Equation

- Can be solved iteratively exactly based on GNSS

time t(i), Orbit oj and good position xR(t(i))

- Must be solved in repeated parameter estimation,

if and xR(t(i)) are unknown)(, iiR tt




with

jTROPi

jIONijSiR

jijí

iSiRObsi tctcc

tttt ,,,,),

)(~(

~)(

~)(

oxx

andc

tiji

i

)(~

)(''''

2

2

c

U

c

v

2

1

m

m

m

mm

f

f

f

ff

x

x

x

xx

t

t

t

tt

c

tt i

ji

i

)(~""

R

1

thR

1

cc

U

EiE22 )(

4.1 Special and General Relativity in general

4.2 GNSS pseudorange-observation scenario

4. Additionally: Relativistic effects (Special and General Relativity)

and


jijí

iSiRObsi

ijí

c

tttt

)(~

)),)(~

(~

)(~

)(

oxx




jTROPi

jIONijSiR

jijí

iSiRObsi tctcc

tttt ,,,,),

)(~(

~)(

~)(

oxx

5. Final Rough Pseudo-Range observation equation

)()()))(sin()((

)(())()(()((

)),)(~

(~

)(~

)()(

,,,

,

)(~

ijTROPii

jIONi

j

i2ijS

iiREi

jSi

iR2iiR


jijí

iSiRObsiObsi

tttEeac

2 ttc

)tttc

1ttc

c

ttttt

ijí

xωxx

oxx

Unknowns: )(),(),,( , iiRiR tttzyx

Corrections: Sagnac Ri, Relativity Sj Clock Sj

Further Corrections, Add. rel. clock variations, Add. ModelsION, TROP, Earth Dynamics

Observations:

)( iiR tt and

ObsjiObsi

ji ct )(



Pseudorange Modeling in ECEF and GNSS-timejTROPi

jIONijSiR

jijí

iSiRObsi tctcc

tttt ,,,,),

)(~(

~)(

~)(

oxx

Further Clock/Time-Bias

• GNSS differential code biases (DCB)

DCB products

available from

01/2013

IGG – updated

daily (daily

interval)

DLR – updated

quarterly (both

weekly and

daily intervals)


Phase Modelling in OPPP / PPP-K

“Wind-up

(“Phase wind up”)

)()(~)()( ,,,,jiji

jTROPi

jIONijSiRi

jí

t at ns"Observatio-Phase" as Stored

ObsijiObsi

ji NtctcttDt

Ri,

Uncalibrated phase delays

(UPD) relevant to undifferen-

ted integer fixing for PPP

//

Cm-accurate PPP

1. PPP (float): No ambiguity fixing → Long convergence time (hours)

2. PPP (fixed): Amb.-Fixing without further local information on

ionosphere and troposphere: 15 or more min.

3. PPP- RTK

3.1 Local information (e.g. Trimble Centre-Point-RTX, QZZS-CLAS)

3.2. Lin. Combinations (WL-NL(via L3), Melbourne-Wübbena,

Code-Phase-LK (GALILEO))

3.3 Ambiguity Function Methods (eliminates also Cycle Slips)


)dt)ff(int(])cyc['int(Dit

0t

tSRii Cycles Di at Phase-

Measurements. Error: „Cycle-

Slip(s)“

Doppler-Frequency Measurement Types

)(c

ff ff RS

RS

RSt

StR

tS rr

rr

rr

2.) General Measurment for

Positioning and Navigation

1.) Add. Measurement for

Positioning by Phase-Measurements

)(2

2

c

U

c

v

2

1ff


RTKLIB - Open Source Software (DGNSS/RTK, GNSS-PPP, Postprocessing

Open Source

DGNSS,

PPP-K


GNSS Raw-Data on Smartphone => New Age, in 2017/2018


• Measurement Modes:I. Single mode.

II. DGNSS (e.g. SAPOS).

III. Kinematic.

IV. PPP-K.

• Raw data/RTCM corrections

• Connection through

Bluetooth, TCP/IP &

NTRIP client.

• Database for Object

management.

• Visualization through a map.

• Geocentric, Geographic &

UTM coordinates.

• (a), (b),(c) &(d) shows the

process of measuring and

visualizing a point in single

mode.

(a) (b)

(c) (d)


NAVKA Smartphone-RTK Software and Hardware

Variant 1.1 Pure DGNSS/PPP-K


DGNSS



Variant 1.1 Pure GNSS / DGNSS

DGNSS

OSR-

corrections

1-2 cm for

2 Frequency

Receivers,

1-3 cm for

1-Frequency-

Receivers.

Longer time

for fixing the

ambiguities


PPP-S/-K


• After 15 min: accuracy is within 20 cm

• After 60 min: accuracy is in the sub-cm level

6h measurement:• standard deviation:

• horizontal: 4.0 mm

• vertical: 2.0 mm


Variant 1.1 Pure GNSS PPP-K with Precise SSR (L1/L2 GPS)

convergence times



Variant 1.1 Pure GNSS, PPP-K and IGS - RTService

Results

KINEMATIC PPP (REAL

TIME – IGS/RTS

L1/L2 receiver (blue)

With Outliers!

Post-processed (green)

At present – replacing

IGS/RTS:

Predicted Orbit and Clock

File - recomputed at 1 Hz

with Bernese Software

In test phase

Realtime Results with L1/L2 GPS


Smartphone Data and Antenna Problem


GNSS Raw Data Available on Smartphones

http://gpsworld.com/innovation-precise-positioning-using-raw-gps-measurements-

from-android-smartphones/

http://gpsworld.com/innovation-theres-an-app-for-that/

http://gpsworld.com/innovation-precise-positioning-using-raw-gps-measurements-from-android-smartphones/

http://gpsworld.com/innovation-theres-an-app-for-that/


Position estimates using pseudorange, Doppler and carrier-phase data. The plotted

estimates are differences with respect to the mean values in each component


GNSS Raw-Data on Smatphone – Risks for PREGON-X Variants 2, 3

Ref.:

GPS

World

http://gpsworld.

com/innovation-

precise-

positioning-

using-raw-gps-

measurements-

from-android-

smartphones/

Does it make sense to develop algorithms & software round not appropriate

patch-antennas hardware ?

http://gpsworld.com/innovation-precise-positioning-using-raw-gps-measurements-from-android-smartphones/



GNSS Raw-Data on Smatphone „Mobile GIS“ Leica Geosystems

L1-based: 40 cm Accuracy ~. 7000.-

EUR

L1/L2 based 5 cm ~ 9000.- EUR

Garmin GPSMAP

64st GPS-

/GLONASS-

Navigationsystem

Turn from the

Patch antenna to a

„Retro“ Helix GNSS

Antenna for PPP-K

…... Just a Matter of

Time ?!


Georeferencing and Standards


Seamless Out-/Indoor Georeferencing and

Interoperability based on ITRF/ETRF89

http://inspire-geoportal.ec.europa.eu/


Transformation Concepts

http://inspire-geoportal.ec.europa.eu/


COPAG_DB

DFLBF_DB

1.)

Transforma-

tions-Para-

meters

&

2.) Residuals

Brasil

www.geozilla.de

3D-Datumtranformation Problem - Solution following Molodenski

FEM-Meshes - Regional 7/8 Parametersets. Continuity Conditions)

Genauigkeitsfläche


Transformation -


Height-Referencing and Geoid/Quasi-Geoid Computation

PREGON-X Smartphone for Mobile GIS & Geo-Referencing

Latvia 2018

Monoglia 2018

www.dfhbf.de

http://www.dfhbf.de/


txR 1t,ITRFyy1zyx

1t,zzITRF1 )t(),,()m1()t(x

)tt( )t()t()m( )t()t( 12t.ITRFzz1)j(Pt.ITRFzz1t.ITRFzz1t.ITRFzz2 1112 xRtxRxx

2.7cm/y => 70 cm

(ETRF89 – ITRF)




Using Reference Tranformations to compute grids dynamically

„Dynamic Grid“

„Combined Message Generation“

[(B,L,h)ITRF-related]i [(B,L)T , HT or N)]i

[(B,L,h)ITRF-related]i [(B,L,h)GNSS,ITRF]i

Part 1 - Plate Models

Part 2 - Standard Reference Transformations

Virtual Fitting-Points

Virtual Fitting Points

Dynamic Message Set up by local 7PT Gridding

RTCM 3.x Transformationmessages (2D/1D)

RTCM WG: 2004-2007

http://geozilla.de/files/110-2008-SC104-508.pdf



http://geozilla.de/files/110-2008-SC104-508.pdf




- Extension of NTV2 Standard for Transformation form ITRF

to Classical Datum Systems to “NT3D” -



Transformation Concepts and RTCM-Questions

1.) Use of RTCM Server, e.g. GZTRaS, enabling Dynamic Grid Generation for 2D and 1D

2.) PREGON-X: Extension of NTV2 (2D only) to “NT3D” – 2.1) 2 Steps with

Smartphone Databases for both steps or a Dynamic NT3D-Grid as an Online Service

3.) Who minds RTCM standard further development? Signals ? SSR-Messages?

Time Shedules ?


Further Information: www.navka.de

Book on Parameter-Estimation including Navigation 10/2018https://www.amazon.de/Klassische-robuste-Ausgleichungsverfahren-

Ausbildung-Geoinformatikern/dp/3879076154


https://www.amazon.de/Klassische-robuste-Ausgleichungsverfahren-Ausbildung-Geoinformatikern/dp/3879076154

Geodetic Challenges on Precise Positioning & Navigation ... · GNSS Raw-Data on Smartphone => New...

Documents

Transcript of Geodetic Challenges on Precise Positioning & Navigation ... · GNSS Raw-Data on Smartphone => New...