Research on Intelligent Navigation Algorithm of Long and Short Term Memory Network Based on Firework Algorithm Optimization in Satellite Blocking Environment

This paper proposes an intelligent navigation algorithm based on long and short term memory network (LSTM) optimized by the fireworks algorithm (FWA) to address the issue of low positioning accuracy of pure inertial navigation systems under satellite blocking. The LSTM network is used to provide simulated satellite navigation positioning information, and the training, predicting, and validation modes are designed to evaluate the network's prediction accuracy and state under dynamic changes in the carrier movement environment and measurement conditions. The FWA is utilized to dynamically adjust the LSTM network parameters and maintain pseudo location availability in the shortest possible training time. In the presence of satellite rejection, the FWA-LSTM network is flexibly selected to correct the inertial navigation system and maintain network availability. Simulation results demonstrate that the FWA-LSTM method provides supplementary support for satellite navigation under complex conditions, enhancing the training efficiency and availability ratio of the navigation system.