In view of the combination of sage-husa and strong track filtering (STF) algorithm still existing negative definite noise variance, failing to estimate process and measurement noise simultaneously and STF depending on the measurements excessively, proposed a novel approach to solve the problems stated above. In a newly scalar sequence processing way, discussed and modified the convergence criterion. When in its convergence, only updated measurements noise variance under the control of a improved forgetting factor; when in its divergence, updated priori estimate error covariance and measurements noise variance simultaneously to make innovation sequence orthogonal and have same order of magnitude. The experiments results illustrated that compared with the conventional one, the proposed method could hold the noise variance positive and came to convergence more quickly with higher accuracy. Although the noise variance increased sharply when in the state of divergence, it met our design objective. © 2011 Springer-Verlag Berlin Heidelberg.
CITATION STYLE
Xu, T., Liu, X., & Yang, X. (2011). Improved adaptive algorithm for ship trajectory estimation. In Advances in Intelligent and Soft Computing (Vol. 124, pp. 209–215). https://doi.org/10.1007/978-3-642-25658-5_25
Mendeley helps you to discover research relevant for your work.