Applying covert sonar tracking using low probability of intercept (LPI) sonar systems to the torpedo attack problem where a torpedo tracks and intercepts a target has the potential to improve attack success. Using realistic simulations of the ocean environment and acoustic interactions, we show that the maximum likelihood-probabilistic data association (ML-PDA) algorithm is well suited to this problem. Previous ML-PDA implementations used a multi-pass grid search to find the maximum of the log likelihood ratio. Use of either a genetic algorithm or a newly developed "intelligent search" algorithm is shown to be significantly more efficient making possible real-time implementation. Reliable tracking down to 5dB (postprocessing SNR: in a resolution cell) using a 10-frame sliding window is demonstrated, this being 3.5 dB better than the multi-pass grid search. The ML-PDA algorithm is also shown to be effective in tracking the target through realistic maneuvers
[1]
David E. Goldberg,et al.
Genetic Algorithms in Search Optimization and Machine Learning
,
1988
.
[2]
Thiagalingam Kirubarajan,et al.
Adaptive early detection ML/PDA estimator for LO targets with EO sensors
,
2000,
SPIE Defense + Commercial Sensing.
[3]
Yaakov Bar-Shalom,et al.
Track formation with bearing and frequency measurements in clutter
,
1990
.
[4]
Saleem A. Kassam,et al.
Analysis of CFAR processors in homogeneous background
,
1988
.
[5]
Y. Bar-Shalom,et al.
Low observable target motion analysis using amplitude information
,
1995,
Proceedings of 1995 American Control Conference - ACC'95.