Most Probable Data Association with Distance and Amplitude Information for Target Tracking in Clutter
In this paper, a new target tracking filter combined with data association called most probable and data association (MPDA) is proposed and its performance is evaluated by a series Monte Carlo simulation tests. The proposed MPDA method utilizes both distance to the predicted target measurement and amplitude information as the probabilistic data association with amplitude information (PDA-AI) method however, it is one-to-one association of track and measurement. All the measurements inside the validation gate at the current sampling time are lined up in the order of the closeness to the predicted target measurement. Probability that the measurement is target-originated is evaluated for each measurement by utilizing order statistics. The measurement with the largest probability is selected to be target-originated and the measurement is used to update the state estimate with the probability used as a weighting factor of the filter gain. To accommodate the probabilistic nature of the MPDA algorithm, a filter structure is developed. The proposed MPDA algorithm can be easily extended for multi-target tracking.
KeywordsTarget tracking Data association PDA MPDAF Clutter
Unable to display preview. Download preview PDF.
- 4.C. R. Li and X. Zhi. (1996). PSNF: a refined strongest neighbor filter for tracking clutter. In Proceedings of the 35th CDC, Kobe Japan, December 1996, pp. 2557–2562.Google Scholar
- 6.T. L. song, Y. T. Lim, and D. G. Lee. (2009). A probabilistic strongest neighbor filter algorithm for m validated measurements. IEEE Transactions on Aerospace and Electronic Systems, 45(1).Google Scholar
- 9.T. L. Song and D. S. Kim. (2006). Highest probability data association for active sonar tracking. In Proceedings of the 9th International Conference on Information Fusion, Florence, Italy.Google Scholar
- 12.P. Willett, R. Niu, Y. Bar-Shalom. (2001). Integration of Bayes detection with target tracking. IEEE Transactions on Signal Process, 49(1): 17–29.Google Scholar