SPACE APPLICATIONS OF GNSS

Transcription

1 SPACE APPLICATIONS OF GNSS Penina Axelrad Colorado Center for Astrodynamics Research University of Colorado Boulder International Technical Symposium on Navigation and Timing Nov 2016 Ecole Nationale de l'aviation Civile, Toulouse, France 1

2 SPACE APPLICATIONS of GNSS Onboard Position & Time Precise Orbit Determination Formation flying Remote sensing Attitude Determination Direct, real-time solution for pos, velocity, and time using pseudorange (and possibly Doppler and carrier phase). Supports real-time onboard operations. Batch or filter-smoother processing of pseudorange and carrier phase measurements with improved GPS orbits & clocks or ground station data. Supports operations planning and primary mission payload. Relative positioning of 2 or more vehicles, possibly in real-time for autonomous control. Occulted and reflected GNSS signals used to observe ionosphere, atmosphere, and Earth surface. Single antenna (SNR-based) or multiple antennas (phase-based) on the vehicle. 2

3 SATELLITE ONBOARD PNT GNSS position and time are used onboard for Ground contact & command execution scheduling (~km) Geo-referencing of science data and imagery (~m) Formation flying, approach and rendezvous (decimeter) Orbit maintenance (100m) Support for GNSS-RO and GNSS-R observations (dm to m) Real-time onboard solutions Typically rely on pseudorange (+carrier phase) Point positioning + orbit propagation, or Kalman Filter with simplified dynamic models PPP or RTK corrections can dramatically improve to <10cm 3

4 LEO FEATURES/CHALLENGES Many GNSS visible typical High Doppler & Doppler rate Rapidly changing geometry affects acquisition, tracking, time tagging, navigation data collection Less ionosphere (except very low satellites) No troposphere (except occultations) Radiation environment User orbital motion quite predictable Antenna placement challenges visibility & multipath Onboard processing limitations 4

5 LIMITING FACTORS FOR LEO ONBOARD PNT GNSS orbit & clock predictability is key limiting factor, - Improvements in precise orbit predictions - Improved clock stability - Real-time correction service w/ppp* shown by Hauschild et al. to be capable of better than 10cm 3D positioning, somewhat degraded for poor GPS clocks Available power limits continuous tracking - Lower power devices becoming available - Snapshot approach using direct positioning *Precise Onboard Orbit Determination for LEO Satellites with Real-Time Orbit and Clock Corrections Hauschild, A. et al. Presented at ION GNSS , Portland, OR, Sept

6 PRECISE ORBIT DETERMINATION High precision, dual frequency receiver recording carrier phase and pseudorange data Worldwide network of ground stations (International GNSS Service) Software Batch or Kalman filter High order gravity field, drag models, SRP, attitude Ambiguity resolution, antenna phase center modeling Coordinate frames, tidal loading Tropospheric zenith delay parameters Reduced dynamics force/acceleration parameters estimated Precise GNSS orbit & clock estimates Solve in parallel or Use precise orbits in PPP 6

7 CUBESATS & NANOSATS GPS on cubesats/nanosats is challenging due to size, weight, power, and $ constraints Antenna placement / attitude control Power/battery limits continuous operation Comm limits affect downlinking raw measurements, upload of GNSS eph & clock Onboard computing limits tracking/orbit prop Potential interference from comm & onboard computer systems Trade spaces: GNSS orbits & clocks from GNSS, SBAS, IGS, PPP service Acquisition & tracking - continuous tracking or fast acquisition, AGPS Frequencies number of carriers (iono), bandwidth Onboard propagation/estimation coordinate frames, perturbations 7

8 Radial SNAPSHOT GNSS Time, orbit, and attitude Capture short window of GNSS band(s) Acquire multiple satellites simultaneously by searching position domain rather than delaydoppler space Estimate attitude from SNR Minimize onboard computation for orbit propagation Estimate large timing errors from multiple snapshots In-Track Accuracy depends on quality of orbit & clock predictions uploaded from the ground 8

9 HEO/GEO GEO & GTO operations, science & exploration Large dynamic range weak/strong signals Antenna gain, visibility & multipath Dynamics well known, accuracy requirements are not stringent (10-100m) Few satellites within main lobe Drawing courtesy M. Moreau 1982 GPS in GEO analyzed by P. Jorgenson 1990 s Transponder in GEO tracked on the ground 1997 TEAMSAT, Equator-S, Falcon Gold 2000 AMSAT-OSCAR 40 (AO-40) AO-40 Spacecraft HEO/GEO PNT relies on: Weak signal tracking, main + side-lobes 30+ GPS/GNSS satellites Stable onboard clock 9

10 SPACE SERVICE VOLUME (SSV) Current GPS SSV specs only include main lobe signals ( deg) On-orbit data show significant PNT improvements with main + side lobes Specify signal availability, received power, pseudorange accuracy Courtesy F. Bauer 10

11 GPS BLOCK IIR & IIR(M) ANTENNA PATTERNS, and all 8 IIR-M SVs igurations on the panel Data publicly released by Lockheed Martin and USAF at ed to L1 and L2 for IIR Panel-Pattern-Marquis-Aug2015-revised.pdf wer due to higher power rther increased on IIR-M with L1M, L2M Improved Antenna Panel (7) Marquis, Willard, "The GPS Block IIR Antenna Panel Pattern and its Use on- Orbit," Proceedings of the 29th International Technical Meeting of The Satellite Division of the Institute of Navigation (ION GNSS+ 2016), 11 Portland, Oregon, September 2016, pp

12 hly eccentric orbits set a record the highest-ever reception of rth Radii (Re) Orbit Measured signals and onboard navigation solutions by an Signal operational Availability Performance GPS receiver Contributed of in space by MMS Side Lobes with m) HEO APPLICATION Side At the lowest (Assumes Lobe Magnetospheric point of Signal 24 the Satellite MMS orbit Availability Constellation) Navigator Multi-Scale (MMS) ogee to 25 Re set a record as the fastest operational GPS From Update L1 Signal on GPS Availability Modernization receiver for in Space space, Operations Main velocities & Science Lobe over Missions, Only 35,000 Progress on Enhancing Main and the GPS Side Space Lobes Service Volume (SSV), Joel J. km/h Signal K. Parker Availability Contributed by Side Lobes 4 or More SVs Visible (Assumes 24 Satellite Never Constellation) 99% L1 1 or Signal More Availability SVs Visible Main Lobe 59% Only Main and Side Lobes 100% 4 or No More SVs SVs Visible Never 41% 99% Never Current 1 or More Spec: SVs L1 Visible Signal Availability 59% 4 or more SVs visible: 100% >1% No SVs Visible 41% Never RED radius CYAN - # tracked Recent Flight Data From Magnetosphere Multi-Scale (MMS) Mission Current Spec: L1 Signal Availability 4 or more SVs visible: >1% Recent Flight Data From Magnetosphere Multi-Scale (MMS) Mission Signal strength (C/N0) vs. position in orbit Signal strength (C/N0) vs. position in orbit 14 Curre Four PRs Current spec: availa Four or more than o PRs shall be 1% o available more than or equal to 1% of the time GEO M 4 satss 1 MMS is seeing 100% 12 15

13 How GNSS plays a key role in remote sensing System, data, and products are available globally 13

14 GNSS REMOTE SENSING GNSS reflections & occultations to study Earth surface & atmosphere. Characterizing reflections delay, attenuation, distortion Characterizing occultations contribution of platform, receiver, antenna, and signal properties 14

15 GNSS-RO (RADIO OCCULTATIONS) KEY COMPONENTS & ALGORITHMS ~Upward pointing antennas used to observe GPS for POD Limb-pointing antennas measure signal amplitude and or more as they pass through iono & tropo ~ 100 sec Precise geometric range removed to construct excess phase Excess phase/doppler shift mapped to bending angle Abel inversion maps bending angle to refractivity Refractivity related to temperature, pressure, water vapor cosmic.ucar.edu 15

16 GNSS-RO SATELLITES FORMOSAT 3/COSMIC (6) GPS/MET, C/NOFS, CHAMP, SAC-C, GRACE, Metop, TerraSAR-X Future COSMIC 2 (6/12), scientific commercial constellations secondary, hosted payloads FORMOSAT-7 Occultations 3 Hrs Coverage

17 GPS-RO EXAMPLE Excess Doppler (blue) and SNR (red) as a function of straight line height Data from 17

18 IONOSPHERIC SENSING / SPACE WEATHER Ground-based or space-based TEC GNSS-RO Global Iono Models from Komjathy et al,

19 GNSS REFLECTOMETRY (GNSS-R) GNSS Satellites PRN coding (C/A, P(Y)) (GNSS bistatic radar) Retreive Sea surface winds Sea ice Soil moisture Altimetry surface height Reflected signals are Delayed Attenuated Broadened Direct Signal RHCP Range cells 19

20 Surface effects on signals Relative to the direct signal, the reflected signal is: Delayed, Attenuated, Broadened Delay (Altimetry) Correlation Power Bistatic Cross-section (Winds & Soil Moisture) Increasing Roughness Direct Signal Reflected Signal Delay (range) 20

21 CYGNSS NASA EV2 Mission Launch Scheduled Dec 12, 2016 Science Goal is to understand the coupling between ocean surface properties, moist atmospheric thermodynamics, radiation, and convective dynamics in the inner core of a Tropical Cyclone. Prof. C. Ruf 21

22 SSTL TechDemoSat-1 SGR-ReSI TechDemoSat Launched in July 2014 Sun-synchronous orbit 650 km altitude Space GNSS Receiver Remote Sensing Instrument Predict specular delay & open loop map Accurate/repeatable amplitude needed Download DDM values, averaged ~1s 13 dbi nadir-pointing antenna, L1 only Surrey Satellite Technology Ltd. Delay Dopper Map Data from: M. Unwin, et al., "Spaceborne GNSS-Reflectometry on TechDemoSat-1: Early Mission Operations and Exploitation," in IEEE JSTARS, vol. 9, no. 10, pp , Oct Surrey Satellite Technology Ltd. 22

23 GNSS-R Altimetry from TDS Retrack signal in the DDM either 70% point or highest slope Predict geometry from precise GPS orbits, TDS orbit solutions in metadata, specular point on the earth DTU10 mean sea surface model Tropospheric model UNB3 model Ionospheric model for LEO Montenbruck/Gill 2D global TEC map from JPL Thin layer model with altitude-dependent effective height for scaling Baseline offset between zenith and nadir antennas 23

24 GNSS-R ALTIMETRY Rx H DTU Incident Rays The signal path delay (δ) is measured and predicted and a height is calculated H DTU = δ DTU / 2 sin(θ) H measured = δ measured / 2 sin(θ) ΔH H measured H DTU WGS84 H measured θ Local Level DTU10 Actual Surface 24

25 GNSS-R ALTIMETRY Example results using data from SSTL TechDemoSat SGR-ReSI Mean Error of -4.8m 1s σ=7m 10s σ=3m Performance probably limited by knowledge of orbit (a) This accuracy not at the level of active radar or laser altimeters Figure 2: (a) Measured surface t opography wit h respect t o the WGS84 reference 25 ellip (b) DT U10 [10] mean sea surface model shown wit h respect t othewgs84 reference ellip (b)

26 Initial Results - Height Error wrt DTU10 MSS Mean Error of -4.8m 1s: σ=7m 10s: σ=3m Mean error consistent across data sets, can be calibrated Variability not yet at the level of tracking noise Work in-progress 26

27 GNSS-R SURFACE MAPPING Ocean Ice Ice shown on image from MASIE-NH (NIC) GPS ice detection based waveform width 70% power Red < 190m, Blue > 190 m 27

28 GNSS IMPROVEMENTS FOR SPACE Clocks Long and short term stability Ensemble, failure detection Orbits attitude, maneuvers, accelerometers (SRP?) Transmitter Characteristics Temperature, antenna pattern, Interchannel biases, power levels Receivers Stable clocks to reduce phase noise & allow longer integration Amplitude measurements Interference monitoring Integrated 6DOF motion Knowledge of environment Efficient processing of multitude of signals Signals of opportunity How do we create new infrastructure that expands global navigation/timing 28

29 Acknowledgments & References Analysis of the TechDemoSat data at CU supported by NASA JPL RSA No Background information on GPS SSV was provided by J.J. Miller, F. Bauer, and A.J. Olia. Hauschild, A. et al., Precise Onboard Orbit Determination for LEO Satellites with Real-Time Orbit and Clock Corrections, ION GNSS , Portland, OR, Sept R. Leandro, M. Santos, and R. B. Langley, UNB Neutral Atmosphere Models: Development and Performance, ION GPS2006, Monterey, CA, United States, 2006, pp Marquis, Willard, "The GPS Block IIR Antenna Panel Pattern and its Use on-orbit," ION GNSS , Portland, OR, Sept J.J. Miller, et al., Achieving GNSS Compatibility and Interoperability to Support Space Users, ION GNSS , Sept Montenbruck, O. and E. Gill, Ionospheric Correction for GPS Tracking of LEO Satellites, Journal of Navigation, vol. 55, no. 2, pp , 2002/05. Montenbruck O., Steigenberger P., Khachikyan R., Weber G., Langley R.B., Mervart L., Hugentobler U., IGS-MGEX: Preparing the Ground for Multi-Constellation GNSS, Science InsideGNSS 9(1):42-49 (2014). Moreau, Autonomous Navigation in High Earth Orbits, PhD Dissertation, University of Colorado, J.K. Parker, Update on GPS Modernization for Space Operations & Science Missions, Progress on Enhancing the GPS Space Service Volume (SSV), Unwin, M et al., "Spaceborne GNSS-Reflectometry on TechDemoSat-1: Early Mission Operations and Exploitation," in IEEE JSTARS, vol. 9, no. 10, pp , Oct doi: /JSTARS Unwin, M. et al., The SGR-ReSI - A New Generation of Space GNSS Receiver for Remote Sensing, ION GNSS 2011, Portland, USA, 2010, pp L. Winternitz, et al., Global Positioning System Navigation Above 76,000 km for NASA s Magnetospheric Multiscale Mission, 2016 AAS GN&C Conference; 39th; 5-10 Feb