Estimation with Applications to Tracking and Navigation
Estimation with Applications to Tracking and Navigation
Non-linear matched filtering for object detection and tracking
Pattern Recognition Letters
Particle filters for maneuvering target tracking problem
Signal Processing
Hi-index | 0.08 |
Due to the measurement time-delay, noises of the charge coupled device (CCD) tracker and maneuvering target model uncertainty characteristic in opto-electronic tracking system, the current statistical model is utilized to describe the unknown acceleration of the maneuvering target. H^~ filter algorithm based on acceleration non-zero mean value time-correlation model is adopted for state prediction of opto-electronic tracking maneuvering targets. Some simulation and experimental results show that the prediction precision by the proposed method is nearly to be twice as much as Kalman filtering. It indicates that the H^~ filter is effective in overcoming model change and stochastic measurement noise, and has high accuracy and good robustness.