The sudden change of vehicle driving environment will reduce the positioning accuracy of single sensor for target vehicle. In this paper, strong tracking unscented Kalman filter (STUKF) algorithm is used to fuse vision and radar sensors to realize multi-sensor fusion positioning of target vehicle and improve positioning accuracy. Firstly, the vehicle motion model is established, and the unscented Kalman filter (UKF) algorithm is studied. In order to improve the robustness of the UKF algorithm, a strong tracking filter (STF) is introduced into the UKF calculation process, and the fading factor is used to correct the prior error covariance in real time to improve the positioning accuracy of the target vehicle. Finally, simulation verifies that the proposed algorithm has good positioning accuracy and robustness under strong transient conditions.
Research on Vision/Radar Sensor Fusion Positioning Algorithm Based on Strong Tracking Filter and Unscented Kalman Filter
12.10.2022
1599397 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Unscented Schmidt–Kalman Filter Algorithm
AIAA | 2015
|BASE | 2014
|Navigation Integration Using the Fuzzy Strong Tracking Unscented Kalman Filter
Online Contents | 2009
|Vehicle Fusion Tracking Based on Square Root Unscented Kalman Filter
British Library Online Contents | 2005
|Unscented Schmidt-Kalman Filter Algorithm
Online Contents | 2015
|