Polarized skylight/MIMU/geomagnetic orientation algorithm based on adaptive Kalman filter

This paper presents an adaptive Kalman filter algorithm for polarized skylight/geomagnetic/GNSS/MIMU bionic integrated navigation system. In the bionic navigation system, the magnetometer, polarized light camera, and satellite antenna are susceptible to environmental interference, which affects the accuracy of the navigation solution. In this paper, a algorithm based on Sage-Husa Kalman filter is given for navigation. The algorithm uses a polarized light camera and a magnetometer to obtain the heading angles information, a satellite receiver to collect speed and position information, and use these messages to estimate and adjust the filter parameters to get an optimal estimate of the system heading angle. In this paper, two land vehicle tests were conducted, and the heading angles were increased by 20.49% and 9.59%, respectively, which proved the effectiveness and availability of the given method.