In this paper, we present a novel and efficient track-before-detect （TBD） algorithm based on multiple-model probability hypothesis density （MM-PHD） for tracking infrared maneuvering dim multi-target. Firstly, the ...In this paper, we present a novel and efficient track-before-detect （TBD） algorithm based on multiple-model probability hypothesis density （MM-PHD） for tracking infrared maneuvering dim multi-target. Firstly, the standard sequential Monte Carlo probability hypothesis density （SMC-PHD） TBD-based algorithm is introduced and sequentially improved by the adaptive process noise and the importance re-sampling on particle likelihood, which result in the improvement in the algorithm robustness and convergence speed. Secondly, backward recursion of SMC-PHD is derived in order to ameliorate the tracking performance especially at the time of the multi-target arising. Finally, SMC-PHD is extended with multiple-model to track maneuvering dim multi-target. Extensive experiments have proved the efficiency of the presented algorithm in tracking infrared maneuvering dim multi-target, which produces better performance in track detection and tracking than other TBD-based algorithms including SMC-PHD, multiple-model particle filter （MM-PF）, histogram probability multi-hypothesis tracking （H-PMHT） and Viterbi-like.展开更多
This paper studies the dynamic estimation problem for multitarget tracking. A novel gat- ing strategy that is based on the measurement likelihood of the target state space is proposed to improve the overall effectiven...This paper studies the dynamic estimation problem for multitarget tracking. A novel gat- ing strategy that is based on the measurement likelihood of the target state space is proposed to improve the overall effectiveness of the probability hypothesis density （PHD） filter. Firstly, a measurement-driven mechanism based on this gating technique is designed to classify the measure- ments. In this mechanism, only the measurements for the existing targets are considered in the update step of the existing targets while the measurements of newborn targets are used for exploring newborn targets. Secondly, the gating strategy enables the development of a heuristic state estima- tion algorithm when sequential Monte Carlo （SMC） implementation of the PHD filter is investi- gated, where the measurements are used to drive the particle clustering within the space gate. The resulting PHD filter can achieve a more robust and accurate estimation of the existing targets by reducing the interference from clutter. Moreover, the target birth intensity can be adaptive to detect newborn targets, which is in accordance with the birth measurements. Simulation results demonstrate the computational efficiency and tracking performance of the proposed algorithm.展开更多
文摘In this paper, we present a novel and efficient track-before-detect （TBD） algorithm based on multiple-model probability hypothesis density （MM-PHD） for tracking infrared maneuvering dim multi-target. Firstly, the standard sequential Monte Carlo probability hypothesis density （SMC-PHD） TBD-based algorithm is introduced and sequentially improved by the adaptive process noise and the importance re-sampling on particle likelihood, which result in the improvement in the algorithm robustness and convergence speed. Secondly, backward recursion of SMC-PHD is derived in order to ameliorate the tracking performance especially at the time of the multi-target arising. Finally, SMC-PHD is extended with multiple-model to track maneuvering dim multi-target. Extensive experiments have proved the efficiency of the presented algorithm in tracking infrared maneuvering dim multi-target, which produces better performance in track detection and tracking than other TBD-based algorithms including SMC-PHD, multiple-model particle filter （MM-PF）, histogram probability multi-hypothesis tracking （H-PMHT） and Viterbi-like.
基金supported by the Aeronautical Science Foundation of China(No.201401P6001)
文摘This paper studies the dynamic estimation problem for multitarget tracking. A novel gat- ing strategy that is based on the measurement likelihood of the target state space is proposed to improve the overall effectiveness of the probability hypothesis density （PHD） filter. Firstly, a measurement-driven mechanism based on this gating technique is designed to classify the measure- ments. In this mechanism, only the measurements for the existing targets are considered in the update step of the existing targets while the measurements of newborn targets are used for exploring newborn targets. Secondly, the gating strategy enables the development of a heuristic state estima- tion algorithm when sequential Monte Carlo （SMC） implementation of the PHD filter is investi- gated, where the measurements are used to drive the particle clustering within the space gate. The resulting PHD filter can achieve a more robust and accurate estimation of the existing targets by reducing the interference from clutter. Moreover, the target birth intensity can be adaptive to detect newborn targets, which is in accordance with the birth measurements. Simulation results demonstrate the computational efficiency and tracking performance of the proposed algorithm.
