Jiming Guo | Wuhan University (original) (raw)
Uploads
Papers by Jiming Guo
There are different ways to construct adaptive Kalman filtering (AKF) algorithms. This paper prop... more There are different ways to construct adaptive Kalman
filtering (AKF) algorithms. This paper proposes an
innovative way to simultaneously estimate the variance
matrix R of the measurement vector and the variance
matrix Q of the process noise vector based on the
variance-covariance component estimation by taking the
advantages of the measurement residuals and the process
noise residuals (Wang, 1997, 2009; Wang et al, 2009)
and the measurement redundancy contribution (Ou,
1989). The core of the novel AKF algorithm lies in the
projection of the system innovation vector into the three
groups of residuals: the residuals of the measurement
vector, the residuals of the process noise vector and the
residuals of the predicted state vector exclusive of the
effect of the process noise. The simulated and real GPS
data in kinematic relative positioning mode were used to
demonstrate the performance of the proposed adaptive
Kalman filter. The results from the simulated datasets
confirm to the simulated variance-covariance
components well. The results from real kinematic GPS
datasets are also provided and discussed.
Proceedings of CPGPS 2010 Navigation and Location Services: Emerging Industry and International Exchanges, Aug 2010
This paper focuses on realization of the variance and covariance component estimation in static ... more This paper focuses on realization of the variance and
covariance component estimation in static relative GPS
positioning for the double-differenced measurements in
sequential least squares. The algorithm presented is based
on the algorithm from [Ou, 1989] through the
incorporation into the sequential least-squares (Wang et
al, 2009; etc.). It is practical and easy to implement, and
computationally efficient. Numerical results from
simulated and real static GPS data in relative positioning
mode are presented and discussed.
Proceedings of CPGPS 2010 Navigation and Location Services: Emerging Industry and International Exchanges, Aug 2010
There are different ways to construct adaptive Kalman filtering (AKF) algorithms. This paper pro... more There are different ways to construct adaptive Kalman
filtering (AKF) algorithms. This paper proposes an
innovative way to simultaneously estimate the variance
matrix R of the measurement vector and the variance
matrix Q of the process noise vector based on the
variance-covariance component estimation by taking the
advantages of the measurement residuals and the process
noise residuals (Wang, 1997, 2009; Wang et al, 2009)
and the measurement redundancy contribution (Ou,
1989). The core of the novel AKF algorithm lies in the
projection of the system innovation vector into the three
groups of residuals: the residuals of the measurement
vector, the residuals of the process noise vector and the
residuals of the predicted state vector exclusive of the
effect of the process noise. The simulated and real GPS
data in kinematic relative positioning mode were used to
demonstrate the performance of the proposed adaptive
Kalman filter. The results from the simulated datasets
confirm to the simulated variance-covariance
components well. The results from real kinematic GPS
datasets are also provided and discussed.
There are different ways to construct adaptive Kalman filtering (AKF) algorithms. This paper prop... more There are different ways to construct adaptive Kalman
filtering (AKF) algorithms. This paper proposes an
innovative way to simultaneously estimate the variance
matrix R of the measurement vector and the variance
matrix Q of the process noise vector based on the
variance-covariance component estimation by taking the
advantages of the measurement residuals and the process
noise residuals (Wang, 1997, 2009; Wang et al, 2009)
and the measurement redundancy contribution (Ou,
1989). The core of the novel AKF algorithm lies in the
projection of the system innovation vector into the three
groups of residuals: the residuals of the measurement
vector, the residuals of the process noise vector and the
residuals of the predicted state vector exclusive of the
effect of the process noise. The simulated and real GPS
data in kinematic relative positioning mode were used to
demonstrate the performance of the proposed adaptive
Kalman filter. The results from the simulated datasets
confirm to the simulated variance-covariance
components well. The results from real kinematic GPS
datasets are also provided and discussed.
Proceedings of CPGPS 2010 Navigation and Location Services: Emerging Industry and International Exchanges, Aug 2010
This paper focuses on realization of the variance and covariance component estimation in static ... more This paper focuses on realization of the variance and
covariance component estimation in static relative GPS
positioning for the double-differenced measurements in
sequential least squares. The algorithm presented is based
on the algorithm from [Ou, 1989] through the
incorporation into the sequential least-squares (Wang et
al, 2009; etc.). It is practical and easy to implement, and
computationally efficient. Numerical results from
simulated and real static GPS data in relative positioning
mode are presented and discussed.
Proceedings of CPGPS 2010 Navigation and Location Services: Emerging Industry and International Exchanges, Aug 2010
There are different ways to construct adaptive Kalman filtering (AKF) algorithms. This paper pro... more There are different ways to construct adaptive Kalman
filtering (AKF) algorithms. This paper proposes an
innovative way to simultaneously estimate the variance
matrix R of the measurement vector and the variance
matrix Q of the process noise vector based on the
variance-covariance component estimation by taking the
advantages of the measurement residuals and the process
noise residuals (Wang, 1997, 2009; Wang et al, 2009)
and the measurement redundancy contribution (Ou,
1989). The core of the novel AKF algorithm lies in the
projection of the system innovation vector into the three
groups of residuals: the residuals of the measurement
vector, the residuals of the process noise vector and the
residuals of the predicted state vector exclusive of the
effect of the process noise. The simulated and real GPS
data in kinematic relative positioning mode were used to
demonstrate the performance of the proposed adaptive
Kalman filter. The results from the simulated datasets
confirm to the simulated variance-covariance
components well. The results from real kinematic GPS
datasets are also provided and discussed.