关键词: BP神经网络, 遗传算法, 航迹预测, 仿真

Abstract: Genetic algorithms are applied to the optimal design of BP neural network. And a predictive model of tracking forecast of moving object based on genetic BP neural network is established. The model overcomes the defects of trapping into local optimal point that ordinary neural network algorithms in the training process are easily to be trapped; and higher accuracy and faster convergence speed are obtained. Finally，the model is analyzed and verified by the filed measurement data. The results show that the neural network forecasting model based on genetic algorithm is better than the single neural network forecasting model in prediction accuracy，which can be used to track forecasting.

Key words: BP neural network, genetic algorithms, tracking forecast, simulation