کد مقاله | کد نشریه | سال انتشار | مقاله انگلیسی | نسخه تمام متن |
---|---|---|---|---|
730353 | 892969 | 2012 | 11 صفحه PDF | دانلود رایگان |

For the last few years, integrated navigation systems have been widely used to calculate positions and attitudes of vehicles. The strapdown inertial navigation system (SINS) provides velocity, attitudes and position information, whereas the global positioning system (GPS) provides velocity and position information. A method using neural network (NN) and wavelet-based de-noising technology is introduced into the SINS/GPS/magnetometer integrated navigation system, because system accuracy may decrease during GPS outages. When the GPS is working well, NN is trained using the velocity and position information provided by SINS as input and the corresponding errors as output. Wavelet multi-resolution analysis (WMRA) is also introduced to de-noise the errors, the desired output of NN. Test results showed that velocity accuracies improved using NN, but other accuracies remained poor. By re-training NN with WMRA, the system accuracies improved to the level of using normal GPS signal. In addition, NN trained with WMRA also improved the approximation to the actual model, further enhancing alignment accuracy.
► We establish a INS/GPS/integrated navigation based on neural network and wavelet.
► Neural network can make the system work seamlessly during GPS outages.
► Wavelets can de-noise the errors of the signal which are the outputs of the NN.
► By re-training NN with WMRA, system accuracies have been improved.
► The method proposed can improve the approximation closer to the actual model.
Journal: Measurement - Volume 45, Issue 4, May 2012, Pages 691–701