Skylight Polarization Pattern Enables Autonomous Positioning: A Solar Position Gradient-Based Method

Polarization positioning is an emerging navigation method that offers the advantages of autonomy and error non-accumulation. The existing polarization positioning methods rely on direct measurements of the solar position. The positioning performance, however, will be degraded in partially cloudy environments, where the solar position measurement bias caused by environment will exist. To address the problem, this paper proposes an autonomous positioning method based on the gradient of solar position and presents a units selection strategy in compound eye polarization sensor. As compared with the existing positioning schemes, the proposed method adopts the solar position gradient for autonomous positioning, thereby suppressing the influence of slow-varying bias on positioning accuracy. To further improve the accuracy of solar position gradient, a polarization units selection strategy based on the relative degree of polarization (RDoP) is proposed. To validate the effectiveness of the proposed method, experiments are conducted in plateau, and plain environments. In the experiments, the minimum root mean square error (RMSE) of the positioning results was 0.105° in longitude and 0.172° in latitude. The minimum mean error was 0.069° in longitude and 0.013° in latitude. The study provides a potential solution for autonomous positioning.