【摘要】 多传感器多目标跟踪系统进行目标状态估计的数据关联问题可以阐述为广义S维分配问题。本文提出了一种基于模糊自适应GA的广义S维分配算法,该算法利用六个模糊控制器对符号编码遗传算法的遗传操作进行自适应控制,并将S维分配问题中的目标代价函数极小化问题作为组合优化问题进行求解,同时结合极大似然方法进行目标识别和目标状态估计。在考虑虚警和漏检前提下,对算法进行了稀疏目标和密集目标两种仿真环境下的Monte Carlo试验,对试验结果进行了对比分析。更多还原

【Abstract】 For multisensor-multitarget tracking systems, the data association problem in target state estimation can be formulated as a generalized S-dimensional (S-D) assignment problem. A novel algorithm is proposed to solve the problem employing a fuzzy adaptive genetic algorithm (GA). This algorithm uses 6 fuzzy logical controllers to adaptively control the genetic operations of a symbol-coded genetic algorithm. By a combinational optimal way, it can find the minimize solution of the objective cost function in the S-D assignment problem, and identify the targets or estimate their states combining with a maximum likelihood method. Monte Carlo experiments are performed in both sparse and dense simulation scenario with false alarm and miss detection. Comparison analysis is illustrated.更多还原