THE MODIFIED VITERBI ALGORITHM IN DETERMINING THE NUMBER OF TARGETS IN THE MULTIPLE TARGET TRACKING
132 viewsKeywords:
Markov chains; Hidden Markov model (HMM); Status; Status values; Observation signs; Observation sign sets; Trace functionsAbstract
In this paper, we present some research results for the MTT (Multiple Target Tracking) problem. Specifically, the approach: Use the Hidden Markov Model HMM (Hidden Markov Model) to identify the target in MTT. To define the target in the data set observed in a noisy environment (with both real and drone targets), the paper used the idea of the Viterbi Algorithm in HMM to determine the hidden part of the model, target part in the set of noisy observations. But MTT only has observed information in the past until the present time, so the reversed variable does not exist, and therefore the algorithm “Forward-Backward” can not apply. In this paper, we give the Forward Algorithm and the Modified Viterbi Algorithm and based on the results that apply to solve the problem of targeting in MTT.
References
[1]. N.T.Hang, “Một thuật toán tối ưu bám quỹ đạo mục tiêu của bài toán quan sát đa mục tiêu trong trường hợp có mục tiêu bị che khuất”, Tạp chí các công trình nghiên cứu phát triển Công nghệ thông tin và Truyền thông, số 01 tháng 09. Tr 46-55, 2019.
[2]. G. David Forney, “The Viterbi algorithm”, International Jour-nal of Pattern Recognition and Artificial Intelligence, 61 (3), pp. 268-278, 1973.
[3]. George Slade, “The Viterbi algorithm demysti”, www.researchgate.net, 2013.
[4]. Zoubin Ghahramani, “An Introduction to Hidden Markov Models and Bayesian Networks”, International Journal of Pattern Recogni-tion and Artificial Intelligence, 15 (1), pp. 9-42, 2001.
[5]. Olivier Cappe, Eric Moulines, and Tobias Ryden, “Inference in hidden Markov models”, Springer Series in Statistics. Springer, New York, 2005.
