Title :
Better M sequence design for parameter identification based on EKF
Author_Institution :
Dept. of Comput. Sci. & Eng., Akita Univ.
fDate :
1/1/1998 12:00:00 AM
Abstract :
The paper discusses a better form of the M sequence input used for the parameter identification based on the EKF. The degree of observability is measured by maximising the expected determinant of the observability Gramian and is employed to get a better M sequence input. The effectiveness of the design is illustrated in a DC servomotor parameter estimation problem. The better M sequence inputs are different from each other, depending on unknown parameters
Keywords :
Kalman filters; nonlinear filters; observability; parameter estimation; sequences; DC servomotor; EKF; M sequence design; degree of observability; extended Kalman filter; observability Gramian; parameter identification;
Journal_Title :
Control Theory and Applications, IEE Proceedings -
DOI :
10.1049/ip-cta:19981581