Combined Navigation Method of RBF Neural Network Based on Quantum Genetic Algorithm in Edge Devices

The combined navigation of Strapdown inertial navigation system (SINS)/Global positioning system (GPS) has been widely applied in various fields currently. Meanwhile, the standard Kalman filtering and the extended Kalman filtering are the widely applied algorithms in the navigation system. However, those algorithms need to be further improved in processing nonlinear signals and uncertain disturbance. Because of the precision of the combined navigation is far from perfect and has a weak robustness. The combined navigation method of RBF neural network based on the quantum genetic algorithm is put forward in this paper. The parameters of RBF neural network are adjusted and optimized using the quantum genetic algorithm. Meanwhile, the Kalman filtering gains are also adjusted through neural network to increase the Signal to Noise ratio (SNR). Thus, the navigation precision and the anti-interference property get improved. In the simulation test about the combined navigation system of SINS/GPS, test results show that the algorithm could realize a better precision and robustness. In addition, the validity of the algorithm is also verified in this paper.