一种改进的联合交互式多模型概率数据关联算法

摘要

In the domain ot multiple-target tracking, because processing to public echoes in the intersection of asso- ciation gates by using of probabihstic data association （PDA） algorithm is too simple, therefore association perform- ante is bad in the environment of dense targets or intersectant targets. Considering the impact of the public echoes on the track updating, the PDA algorithm is improved. An improved combined interacting multiple models probabil- ity data association algorithm （IC-IMMPDA） is proposed by combining interacting multiple model （IMM） algorithm with improved PDA algorithm. The simulation results demonstrate that performance of the new algorithm in the envi- ronment of＂ dense targets or intersectant targets is much better than that of the combined interacting multiple models probabilistic data association algorithm （C-IMMPDA）, and it also features low calculation load; which will make it a good application prospect in practical engineering.%在多目标跟踪领域，由于概率数据关联（PDA）算法对关联门相交区域内公共回波的处理过于简单，因而在目标密集或交叉环境中关联性能差。考虑公共回波对航迹更新的影响对PDA算法进行改进，并将改进的PDA算法和交互式多模型（IMM）算法结合提出了一种改进的联合交互式多模型概率数据关联（Ic．IMMPDA）算法，并在Matlab环境下将该算法与联合交互式多模型概率数据关联算法（C．IMMPDA）算法对比研究。仿真结果表明：在目标密集或交叉的环境下，新算法跟踪性能远好于C—IMMPDA算法，同时保持了较小的计算量，在实际工程中有很好的应用前景。