摘要
针对常规自主式水下航行器(AUV)深度传感器尚未考虑受内部环境温度、电流等影响的问题,基于小波神经网络多传感器信息融合技术,建立了融合温度传感器、电流传感器和深度传感器样本信息的AUV深度测量紧致型小波神经网络结构模型,并利用基于梯度的学习算法进行求解。试验数据表明,深度传感器测量精度经过小波神经网络信息融合后,测量误差大大降低,很好地消除了环境温度、电流等干扰因素对深度传感器测量精度的影响。
To consider the influences of internal ambient temperature, current and other factors on the measurement accuracy of an autonomous underwater vehicle(AUV) depth sensor, a compact type depth measurement model is estab- lished by means of the wavelet neural network based multi-sensor information fusion technology. This model fuses the sample information of temperature, current and depth sensors. The gradient based learning algorithm is adopted to solve the model. Experimental results show that the measurement error of the depth sensor is significantly reduced after wavelet neural network based information fusion, and the influences of the disturbing factors, such as internal ambient temperature and current, on the depth sensor are eliminated. It is concluded that wavelet neural network can be used in the information fusion of AUV depth sensor to effectively improve the performance of the depth sensor.
出处
《鱼雷技术》
2016年第4期267-270,共4页
Torpedo Technology