Abstract: Genetic operator is used to optimize particle swarm optimization algorithm, and to form a multi-source data fusion model based on GA-PSO. The model overcomes the defects of trapping into local optimal point that ordinary particle swarm optimization algorithm easily falls into in the training process, and reaches higher accuracy and faster converg ence speed. Finally, the fusion mode of the target ship track point data is validated by multiple sensors. The results show that track accuracy based on GA-PSO fusion model is better than that detected by every single sensor in detecting the target ship track point data, and the GA-PSO fusion model is more suitable to do ship tracking and forecast

Key words: genetic operators, particle swarm optimization algorithm, multi-source data fusion, MATLAB simulation

摘要： 研究利用遗传算子对粒子群算法进行优化设计,建立了基于遗传算子的粒子群算法多源数据融合模型。该 模型克服了粒子群算法在训练过程中容易陷入局部极值的缺陷,得到了更高的学习精度和更快的收敛速度。利用 多传感器检测到的目标船舶航迹点数据进行了融合验证,MATLAB仿真结果表明,基于遗传算子的粒子群算法融 合模型融合后的目标船舶航迹点比各传感器单独检测到的目标船舶航迹点数据更加精确,更适用于船舶航迹的跟 踪及预测。

关键词: 遗传算子, 粒子群算法, 多源数据融合, MATLAB仿真