哈尔滨工程大学学报2026，Vol.47Issue(4)：819-828,10.DOI:10.11990/jheu.202407020
基于概率数据互联的水下目标检测前跟踪方法
Track-before-detect method for underwater targets based on probabilistic data association
摘要
Abstract
To address the performance degradation of track-before-detect methods for underwater weak target tracking and detection in clutter environments,a track-before-detect method based on data association is proposed.The proba-bilistic data association algorithm is employed for data association,while a particle filter is utilized to compute the tar-get existence probability and estimate the target state.Simulation and lake trial data processing results demonstrate that the proposed method achieves a correct target detection probability of over 92%at an SNR of 6 dB.Compared with traditional track-before-detect methods,the root mean square error of the tracking position is reduced by more than 62%,the tracking convergence speed is improved by over 3 frames,and the correct detection probability is in-creased by more than 10%.The proposed method significantly enhances tracking accuracy and detection probability.关键词
杂波/检测前跟踪/粒子滤波/概率数据互联/联合概率数据互联/目标跟踪/检测概率/状态估计Key words
clutter/track-before-detect/particle filter/probabilistic data association/joint probabilistic data asso-ciation/target tracking/detection probability/state estimation分类
数理科学引用本文复制引用
任聪,李亚林,聂东虎,欧阳哲,温佳伟..基于概率数据互联的水下目标检测前跟踪方法[J].哈尔滨工程大学学报,2026,47(4):819-828,10.基金项目
国家自然科学基金(11974090,11774074) （11974090,11774074）
深圳市科技计划项目(JSGG20220831103800001). （JSGG20220831103800001）
