Data Missing Process by Extended Kalman Filter with Equality Constraints
2017
In this paper, the extended Kalman filter (EKF) is used to estimate the position of the feature points when data missing occurs, taking the feature extraction of plane moving robot ceiling-based positioning as background and the coordinates of the feature points in the image plane as objects. Firstly, the acceleration model of the feature points in the image plane is established as the motion equation, and the motion information of the feature points is extracted by filter. Then, the equality constraints of the feature points are added to the filter scheme to increase the measurement information. In the case where there is a loss of data, that is, the feature points are lost partly, and the predicted values of the lost points are estimated as the true value. By comparing the filtering results, it shows that the addition of equality constraints can not only enhance the filtering effect, but also can estimate the loss points more effectively. Finally, the validity of the filtering scheme is verified by a numerical example.
Keywords:
- Correction
- Source
- Cite
- Save
- Machine Reading By IdeaReader
10
References
0
Citations
NaN
KQI