摘要
针对毫米波雷达虽能获取极坐标下目标数据,但无法直接获取目标物位移分量且受系统噪声影响无法进行高精度定位跟踪的问题,提出一种利用反向传播(BP)神经网络校正扩展卡尔曼算法的无人帆船目标物的精确定位方法。通过扩展卡尔曼算法融合毫米波雷达数据,估计动态目标直角坐标系下的位移分量;并用BP神经网络算法校正扩展卡尔曼滤波器,以降低动态目标物运动的不确定性及噪声影响。仿真结果表明,该方法能够降低系统噪声的影响,更准确地进行目标物的定位追踪。
Aiming at the problems that the millimeter-wave radar can obtain the target data under polar coordinates,but it can’t directly obtain the displacement component of the target object,and it is also unable to carry out high-precise location tracking due to the influence of systematic noise,the paper proposed a precise positioning method of unmanned sailboat target object using an extended Kalman algorithm aided by Back Propagation(BP)neural network:the extended Kalman algorithm was fused with the millimeter-wave radar data to estimate the displacement component of the dynamic target in Cartesian coordinate system,and the BP neural network algorithm was used to correct the extended Kalman filter for reducing the motion uncertainty of the dynamic target and the influence of noise.Simulational result showed that the proposed method could help reduce the influence of system noise,and achieve accurate target localization and tracking.
作者
部德强
BU Deqiang(College of Engineering,Ocean University of China,Qingdao,Shandong 266100,China)
出处
《导航定位学报》
CSCD
2021年第6期84-89,111,共7页
Journal of Navigation and Positioning
关键词
无人帆船目标物
神经网络
扩展卡尔曼
定位追踪
unmanned sailboat target object
neural network
extended Kalman
localization and tracking
