Title :
A novel recursive algorithm for directional forgetting
Author :
Cao, Liyu ; Schwartz, Howard M.
Author_Institution :
Dept. of Syst. & Comput. Eng., Carleton Univ., Ottawa, Ont., Canada
Abstract :
An algorithm for directional forgetting is developed. This algorithm performs exponential forgetting according to the direction of data vector, thus preventing the problem known as estimator windup which is a characteristic of the standard exponential forgetting algorithm. This algorithm is able to track fast parameter changes and is similar in complexity to the standard least square algorithm. The superior performance of the algorithm is verified via theoretical and simulation studies
Keywords :
covariance matrices; recursive estimation; tracking; directional forgetting; exponential forgetting; fast parameter changes; recursive algorithm; Algorithm design and analysis; Drives; Eigenvalues and eigenfunctions; Extraterrestrial measurements; Least squares methods; Parameter estimation; Performance analysis; Systems engineering and theory; Time varying systems; Windup;
Conference_Titel :
American Control Conference, 1999. Proceedings of the 1999
Conference_Location :
San Diego, CA
Print_ISBN :
0-7803-4990-3
DOI :
10.1109/ACC.1999.783584
