Development of spaceborne GPS receiver with real-time orbit determination using unscented Kalman filter

Eun Jung Choi, Jae Cheol Yoon, Byoung Sun Lee, Sang Young Park, Kyu Hong Choi

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

Spaceborne GPS receiver is used for not only a primary tracking system for precision orbit determination on ground but also a core equipment to provide real-time ephemeris for onboard attitude control in many missions. In general, the position and velocity accuracy of GPS navigation solution without dynamic filter has 25 m(1 σ) and 1.0 m/s(1 σ), respectively. However, GPS navigation solution from most of spaceborne GPS receiver has abnormal excursions of about 4 km and 3 m/s from normal error range for space operation. These excursions have a negative effect on the attitude control system. In this research, a spaceborne GPS receiver, which includes real-time orbit determination function using unscented Kaiman filter(UKF), was developed to provide more accurate and robust real-time ephemeris. Dynamical orbit determination is a nonlinear problem where the perturbing factors are not easily modeled. Extended Kaiman filter(EKF) is one of the favorable approaches as the navigation state estimator. However, the divergence due to modeling errors, which EKF highly depends on a predefined dynamic model, is a critical problem. In this research, UKF algorithm, which uses a finite number of sigma points to propagate the probability of state distribution, was implemented for the nonlinear dynamic process modeling. And also, optimal UKF algorithm based on Kaiman minimum variance method was applied fro real-time calculations. 7 th order Runge-Kutta numerical integration was applied for orbit propagation and perturbations due to geopotential, gravity of the Sun and Moon, solar radiation pressure, and atmospheric drag were modeled. GPS navigation solution, which consists of position and velocity, was used as observation for UKF. The spaceborne GPS receiver hardware design is mainly built with the 12 channel L 1 C/A code baseband correlator and 32-bit floating point DSP microprocessor. The performance test of real-time orbit determination using UKF was implemented using the COSMIC navigation solution data in the DSP-based spaceborne GPS receiver.

Original languageEnglish
Title of host publication60th International Astronautical Congress 2009, IAC 2009
Pages2891-2896
Number of pages6
Volume4
Publication statusPublished - 2009 Dec 1
Event60th International Astronautical Congress 2009, IAC 2009 - Daejeon, Korea, Republic of
Duration: 2009 Oct 122009 Oct 16

Other

Other60th International Astronautical Congress 2009, IAC 2009
CountryKorea, Republic of
CityDaejeon
Period09/10/1209/10/16

Fingerprint

orbit determination
Kalman filters
Kalman filter
GPS
receivers
filter
filters
navigation
ephemeris time
attitude control
range errors
COSMIC
geopotential
performance tests
radiation pressure
microprocessors
solar radiation
moon
correlators
numerical integration

All Science Journal Classification (ASJC) codes

  • Space and Planetary Science
  • Astronomy and Astrophysics

Cite this

Choi, E. J., Yoon, J. C., Lee, B. S., Park, S. Y., & Choi, K. H. (2009). Development of spaceborne GPS receiver with real-time orbit determination using unscented Kalman filter. In 60th International Astronautical Congress 2009, IAC 2009 (Vol. 4, pp. 2891-2896)
Choi, Eun Jung ; Yoon, Jae Cheol ; Lee, Byoung Sun ; Park, Sang Young ; Choi, Kyu Hong. / Development of spaceborne GPS receiver with real-time orbit determination using unscented Kalman filter. 60th International Astronautical Congress 2009, IAC 2009. Vol. 4 2009. pp. 2891-2896
@inproceedings{4c54b720877d454aaf2c56ea7e77ed9b,
title = "Development of spaceborne GPS receiver with real-time orbit determination using unscented Kalman filter",
abstract = "Spaceborne GPS receiver is used for not only a primary tracking system for precision orbit determination on ground but also a core equipment to provide real-time ephemeris for onboard attitude control in many missions. In general, the position and velocity accuracy of GPS navigation solution without dynamic filter has 25 m(1 σ) and 1.0 m/s(1 σ), respectively. However, GPS navigation solution from most of spaceborne GPS receiver has abnormal excursions of about 4 km and 3 m/s from normal error range for space operation. These excursions have a negative effect on the attitude control system. In this research, a spaceborne GPS receiver, which includes real-time orbit determination function using unscented Kaiman filter(UKF), was developed to provide more accurate and robust real-time ephemeris. Dynamical orbit determination is a nonlinear problem where the perturbing factors are not easily modeled. Extended Kaiman filter(EKF) is one of the favorable approaches as the navigation state estimator. However, the divergence due to modeling errors, which EKF highly depends on a predefined dynamic model, is a critical problem. In this research, UKF algorithm, which uses a finite number of sigma points to propagate the probability of state distribution, was implemented for the nonlinear dynamic process modeling. And also, optimal UKF algorithm based on Kaiman minimum variance method was applied fro real-time calculations. 7 th order Runge-Kutta numerical integration was applied for orbit propagation and perturbations due to geopotential, gravity of the Sun and Moon, solar radiation pressure, and atmospheric drag were modeled. GPS navigation solution, which consists of position and velocity, was used as observation for UKF. The spaceborne GPS receiver hardware design is mainly built with the 12 channel L 1 C/A code baseband correlator and 32-bit floating point DSP microprocessor. The performance test of real-time orbit determination using UKF was implemented using the COSMIC navigation solution data in the DSP-based spaceborne GPS receiver.",
author = "Choi, {Eun Jung} and Yoon, {Jae Cheol} and Lee, {Byoung Sun} and Park, {Sang Young} and Choi, {Kyu Hong}",
year = "2009",
month = "12",
day = "1",
language = "English",
isbn = "9781615679089",
volume = "4",
pages = "2891--2896",
booktitle = "60th International Astronautical Congress 2009, IAC 2009",

}

Choi, EJ, Yoon, JC, Lee, BS, Park, SY & Choi, KH 2009, Development of spaceborne GPS receiver with real-time orbit determination using unscented Kalman filter. in 60th International Astronautical Congress 2009, IAC 2009. vol. 4, pp. 2891-2896, 60th International Astronautical Congress 2009, IAC 2009, Daejeon, Korea, Republic of, 09/10/12.

Development of spaceborne GPS receiver with real-time orbit determination using unscented Kalman filter. / Choi, Eun Jung; Yoon, Jae Cheol; Lee, Byoung Sun; Park, Sang Young; Choi, Kyu Hong.

60th International Astronautical Congress 2009, IAC 2009. Vol. 4 2009. p. 2891-2896.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Development of spaceborne GPS receiver with real-time orbit determination using unscented Kalman filter

AU - Choi, Eun Jung

AU - Yoon, Jae Cheol

AU - Lee, Byoung Sun

AU - Park, Sang Young

AU - Choi, Kyu Hong

PY - 2009/12/1

Y1 - 2009/12/1

N2 - Spaceborne GPS receiver is used for not only a primary tracking system for precision orbit determination on ground but also a core equipment to provide real-time ephemeris for onboard attitude control in many missions. In general, the position and velocity accuracy of GPS navigation solution without dynamic filter has 25 m(1 σ) and 1.0 m/s(1 σ), respectively. However, GPS navigation solution from most of spaceborne GPS receiver has abnormal excursions of about 4 km and 3 m/s from normal error range for space operation. These excursions have a negative effect on the attitude control system. In this research, a spaceborne GPS receiver, which includes real-time orbit determination function using unscented Kaiman filter(UKF), was developed to provide more accurate and robust real-time ephemeris. Dynamical orbit determination is a nonlinear problem where the perturbing factors are not easily modeled. Extended Kaiman filter(EKF) is one of the favorable approaches as the navigation state estimator. However, the divergence due to modeling errors, which EKF highly depends on a predefined dynamic model, is a critical problem. In this research, UKF algorithm, which uses a finite number of sigma points to propagate the probability of state distribution, was implemented for the nonlinear dynamic process modeling. And also, optimal UKF algorithm based on Kaiman minimum variance method was applied fro real-time calculations. 7 th order Runge-Kutta numerical integration was applied for orbit propagation and perturbations due to geopotential, gravity of the Sun and Moon, solar radiation pressure, and atmospheric drag were modeled. GPS navigation solution, which consists of position and velocity, was used as observation for UKF. The spaceborne GPS receiver hardware design is mainly built with the 12 channel L 1 C/A code baseband correlator and 32-bit floating point DSP microprocessor. The performance test of real-time orbit determination using UKF was implemented using the COSMIC navigation solution data in the DSP-based spaceborne GPS receiver.

AB - Spaceborne GPS receiver is used for not only a primary tracking system for precision orbit determination on ground but also a core equipment to provide real-time ephemeris for onboard attitude control in many missions. In general, the position and velocity accuracy of GPS navigation solution without dynamic filter has 25 m(1 σ) and 1.0 m/s(1 σ), respectively. However, GPS navigation solution from most of spaceborne GPS receiver has abnormal excursions of about 4 km and 3 m/s from normal error range for space operation. These excursions have a negative effect on the attitude control system. In this research, a spaceborne GPS receiver, which includes real-time orbit determination function using unscented Kaiman filter(UKF), was developed to provide more accurate and robust real-time ephemeris. Dynamical orbit determination is a nonlinear problem where the perturbing factors are not easily modeled. Extended Kaiman filter(EKF) is one of the favorable approaches as the navigation state estimator. However, the divergence due to modeling errors, which EKF highly depends on a predefined dynamic model, is a critical problem. In this research, UKF algorithm, which uses a finite number of sigma points to propagate the probability of state distribution, was implemented for the nonlinear dynamic process modeling. And also, optimal UKF algorithm based on Kaiman minimum variance method was applied fro real-time calculations. 7 th order Runge-Kutta numerical integration was applied for orbit propagation and perturbations due to geopotential, gravity of the Sun and Moon, solar radiation pressure, and atmospheric drag were modeled. GPS navigation solution, which consists of position and velocity, was used as observation for UKF. The spaceborne GPS receiver hardware design is mainly built with the 12 channel L 1 C/A code baseband correlator and 32-bit floating point DSP microprocessor. The performance test of real-time orbit determination using UKF was implemented using the COSMIC navigation solution data in the DSP-based spaceborne GPS receiver.

UR - http://www.scopus.com/inward/record.url?scp=77953590493&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77953590493&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9781615679089

VL - 4

SP - 2891

EP - 2896

BT - 60th International Astronautical Congress 2009, IAC 2009

ER -

Choi EJ, Yoon JC, Lee BS, Park SY, Choi KH. Development of spaceborne GPS receiver with real-time orbit determination using unscented Kalman filter. In 60th International Astronautical Congress 2009, IAC 2009. Vol. 4. 2009. p. 2891-2896