[SEMINAR] Computationally Efficient Non-linear Kalman Filters for On-board Space Vehicle Navigation
In this seminar the following topics will be discussed:
- Development of two new UKF based estimation algorithms called the Single Propagation Unscented Kalman Filter (SPUKF) and the Extrapolated Single Propagation Unscented Kalman Filter (ESPUKF) which require only single sigma point propagation for the computation of the sigma points at the next epoch. These sigma points are used to calculate the a priori state and error covariance.
- Theoretical analysis of computation time for the UKF, SPUKF and the ESPUKF is performed. It is shown that the SPUKF and ESPUKF are capable of reducing the processing time of the UKF by upto 90%. The accuracy of the SPUKF is higher than the EKF and the ESPUKF provides accuracy similar to the UKF. The computational advantage of the new lters depends on the complexity of the system, integration step size and number of state elements.
- Establishment of a relation between the degree of non-linearity of system-measurement models and UKF performance. It is observed that there is an upper limit of estimation accuracy improvement by the UKF over the EKF which depends on the sum of the degree of non-linearity of the system and the measurement model. A probabilistic analysis is also performed to show that, for highly non-linear problem, there is a higher probability that the UKF estimation accuracy will be signi cantly better than the EKF estimation accuracy.
- Characterization of e ffect of GNSS Dilution of Precision (DOP) on non-linear Kalman Filter performance for GNSS-based space vehicle navigation.
Sanat Biswas is a PhD student in the School of Electrical Engineering and Telecommunications at the University of New South Wales (UNSW). He received BE (2010) in Instrumentation and Electronics from Jadavpur University and M. Tech (2012) in Aerospace Engineering from Indian Institute of Technology Bombay. Sanat is currently associated with Australian Centre for Space Engineering Research (ACSER) and Satellite Navigation and Positioning (SNAP) Laboratory. He has been awarded the Emerging Space Leaders Grant 2014 by the International Astronautical Federation.