Dynamic-Wave Interference Suppression Based on Angular Increment Assistance for Underwater Imaging Polarization Sensor

Abstract

The imaging polarization sensor (IPS) can provide high-accuracy, robust, and fully autonomous navigation information for underwater unmanned platforms. However, the polarization images in complex underwater scenarios are affected by dynamic-wave interference, severely impacting the accuracy of underwater polarization orientation. Therefore, a suppression method of dynamic-wave interference based on angular increment assistance is proposed. First, the transmission model of polarized light under dynamic-wave interference is constructed, and the error mechanism of the angle of polarization (AOP) is analyzed. Then, combining the inertial angular increment, a total differential increment model of AOP under the condition of carrier motion is established and used as the theoretical prior for the relative change in adjacent frames. On this basis, an AOP deviation estimation method based on complementary filtering is proposed, which exploits the complementary characteristics between the total differential increment of AOP and the actual detected values, achieving dynamic-wave suppression with single-frame input and thereby improving the accuracy of polarization orientation. In actual underwater experiments, the proposed method not only effectively improves the quality of AOP images but also significantly enhances the orientation accuracy. Compared to the existing optimal methods, the underwater static orientation error and the underwater dynamic orientation error are reduced by 26.3% and 33.4%, respectively.