Particle Filter Application Tobearings-Only Tracking
Chinta Mahesh1, M. Prakash Reddy2, S. Koteswara Rao3, Kausar Jahan4
1Chinta Mahesh, Department of ECE, Koneru Lakshmaiah Education Foundation, Vijayawada (A.P), India.
2M.Prakash Reddy, Department of ECE, Koneru Lakshmaiah Education Foundation, Vijayawada (A.P), India.
3S.Koteswara Rao, Department of ECE, Koneru Lakshmaiah Education Foundation, Vijayawada (A.P), India.
4Kausar Jahan, Department of ECE, Koneru Lakshmaiah Education Foundation, Vijayawada (A.P), India.
Manuscript received on 07 March 2019 | Revised Manuscript received on 20 March 2019 | Manuscript published on 30 March 2019 | PP: 374-378 | Volume-8 Issue-5, March 2019 | Retrieval Number: E3161038519/19©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Passive tracking of a target using bearings-only measurement which is carried out in underwater scenario is most widely used. This paper considers the problem of estimating the position and velocity of a target in an underwater scenario. As bearings-only tracking is a non-linear problem, Kalman filter which is linear and traditional filter can’t give correct approximations. The noise in measurement is more as underwater scenario is considered which leads to less accurate estimation. Particle filter (PF) which is a non-linear filter is considered for this problem. In PF, all particles are assigned with weights based on their likelihood computed with respect to obtained measurements. The weights assigned to the particles, after certain time period tend to equal values called sample impoverishment.The main difficulty using PF is sample degeneracy and sample impoverishment. To avoid these problems, different re-sampling techniques can be used, or PF can be combined with other filters like Extended Kalman Filter (EKF), Unscented Kalman filter (UKF), Modified Gain Bearings-only Extended Kalman filter (MGBEKF) etc. In this paper, PF with Systematic re-sampling technique and combined with MGBEKF is considered for analysing the estimation of target parameters. Evaluation of the algorithm is assessed based on the best convergence time of the solution for many scenarios using MATLAB software.
Keyword: Bearings-only tracking, Modified Gain Bearing-Only Extended Kalman Filter, Particle Filter, Signal Processing, Systematic Re-sampling.
Scope of the Article: Big Data Analytics Application Systems