Conical error compensation algorithm based on particle swarm optimization algorithm

For attitude updating algorithm of Strapdown Inertial Navigation System, conical motion is the worst working environment. Due to the use of Taylor's formula, the traditional algorithm has truncation error. As the frequency of vibration increases, the effect of compensation is getting worse and worse, this paper is based on the traditional subsample algorithm, Using the conical compensation residual error square sum as the fitness function, the particle swarm optimization algorithm is used to optimize the depth learning extremum. Taking the initial weights and initial threshold generated by Monte Carlo as input, the average value of multiple optimization results is used as a coning error compensation optimization coefficient, so that the residual error after compensation is the minimum. The simulation results show that compared with the traditional coning error compensation algorithm, neural network and genetic algorithm, this algorithm improves the accuracy of the attitude calculation of strapdown inertial navigation system, and achieves excellent coning error compensation effect.