基于BP神经网络的多传感器数据融合技术优化

传统的数据融合算法要求获得比较精确的对象数学模型,对于复杂的难于建立模型的场合无法适用。为解决上述问题,提出了一种基于BP神经网络算法的多传感器数据融合方法,对对象的先验要求不高,具有较强的自适应能力。仿真结果表明,采用BP神经网络对传感器数据进行融合处理大大提高了传感器的稳定性及其精度,效果良好。

Based on BP neural network multi-sensor optimization of data fusion technology to optimize

Traditional data fusion algorithms need precise malhematical model of objects,which is difficult for modeling complex situations. To address these problems, we propose a neural network algorithm based on BP multi-sensor data fusion method, the object of a priori less demanding, with a strong adaptive ability. Simulation results show that obtained using BP neural network to process sensor data fusion can greatly improve the stability of the sensor and is precision to good effect.