Filter Design for Laser Inertial Navigation System Based on Improved Pigeon-Inspired Optimization

The laser gyroscope of Laser Inertial Navigation System (LINS) eliminates the influence of the locked zone with mechanical dither. The output information of laser gyroscopes must be filtered before use to eliminate vibration noise. Laser gyroscope filters are designed according to the instrument accuracy, calculation capacity, vibration frequency, system dynamic characteristics, and other indicators. In this paper, a pigeon-inspired optimization (PIO) method is proposed for use in filter design. The PIO method can flexibly design filters with excellent performance according to the indicator requirements. In the method, the constraints and indicators of the amplitude, phase and order of the LINS filter are firstly confirmed according to the application requirements; then, the objective function is established, and the parameters to be optimized of the PIO are set according to the order of the filter; finally, the PIO method is used to obtain filter parameters that can satisfy the constraints and achieve better performance. Referring to the idea of biological evolution mechanisms, we propose a new improved pigeon-inspired optimization method based on natural selection and Gaussian mutation (SMPIO), which can obtain more stable results and higher accuracy. In the SMPIO method, the particle swarm is firstly selected by natural selection, that is, the particles are sorted according to the fitness function, and some particles with poor fitness are replaced by those with better fitness; then, all particles are subjected to Gaussian mutation to obtain a better global optimum. SMPIO method can flexibly design filters according to the comprehensive requirements of laser gyro performance and navigation control indicators, which cannot be achieved by traditional filter design methods; the improvement based on natural selection and Gaussian mutation enables SMPIO to have faster convergence speed, and higher accuracy.