摘要
由捷联惯性导航(SINS)和全球定位系统(GPS)组成的组合导航系统在使用时会受周边恶劣环境干扰,导致GPS失锁现象发生。针对GPS失锁后精度迅速下降,无法正确导航的情况,提出了一种基于RBF神经网络辅助组合导航算法。在GPS信号良好且可用时,通过卡尔曼滤波对组合导航输出的导航信息进行数据融合,将解算后的速度和位置信息送入RBF网络进行在线训练;当GPS接收机信号异常导致失锁时,利用训练好的RBF网络补偿载体自身的速度和位置误差信息。该算法可以解决SINS随时间的增加,位置和速度误差逐步累积,导致无法导航的问题。通过跑车实验解算速度和位置信息,结果表明,速度平均误差在0.36m/s以内,位置平均误差在3.14m以内,证实了该算法对组合导航的有效性。
The integrated navigation system,which consists of strapdown inertial navigation system(SINS)and global positioning system(GPS),can be disturbed by the bad surrounding environment,which can lead to GPS outage.In order to solve the problem that GPS navigation can not be done correctly due to the rapid decrease of accuracy after the outage,an integrated navigation algorithm based on RBF neural network is proposed.When GPS signal is good and available,the navigation information output by integrated navigation is fused by Kalman filter,and the speed and position information after computation are sent to RBF network for online training.When the GPS receiver signal is abnormal,the trained RBF network is used to compensate the speed and position error.information of the carrier.This algorithm can solve the problem that SINS position and speed errors accumulate gradually with the increase of time,which makes navigation impossible.The experimental results show that the average speed error is within 0.36m/s and the average position error is within 3.14m,which can prove the effectiveness of the proposed algorithm for integrated navigation.
作者
赵乐宁
李杰
冯凯强
魏晓凯
Zhao Lening;Li Jie;Feng Kaiqiang;Wei Xiaokai(School of Instrumentation and Electronics,North University of China,Taiyuan 030051,China;Key Laboratory of Instrumentation Science and Dynamic Measurement,Ministry of Education,North University of China,Taiyuan 030051,China)
出处
《航天控制》
CSCD
北大核心
2022年第3期37-43,共7页
Aerospace Control
基金
国家自然科学基金(51575500)。
关键词
组合导航
神经网络
误差补偿
GPS
Integrated navigation
Neural network
Error compensation
GPS
