A Method for Assisting Guided Projectile SINS/GNSS Inte-grated Navigation System during GNSS Signal Rejection Based on TOC-NARX

The flight environment of guided projectiles is complex, which can easily lead to signal loss of the global navigation satellite system (GNSS), resulting in the failure of the navigation system composed of the strapdown inertial navigation system (SINS) and GNSS. Aiming to address the above problems, a method of SINS/GNSS integrated navigation assisted by the tornado optimizer with Coriolis force (TOC) and nonlinear autoregressive neural network with external input (NARX) is proposed. TOC is employed to optimize the initial weights and thresholds of NARX to avoid the problem that a single network is easy to fall into local optimal solution and poor generalization. In the network training phase, the specific force, angular velocity, and the position and velocity increments of SINS are utilized as inputs, while the position and velocity increment of GNSS are used as outputs. When the GNSS signal is rejected, the trained network is utilized to predict the GNSS signal, thereby ensuring the normal operation of the navigation system. To validate the method, a series of simulation experiments were conducted. The simulation results demonstrate that the prediction accuracy of this method is significantly improved compared to pure inertial navigation and unoptimized NARX models. This enhancement can support guided projectile integrated navigation during GNSS signal suppression.