This book systematically and comprehensively elaborates on the intelligent information processing technology for a bioinspired polarization compass. The content of this book are briefly consisted of three parts. The research background and significance of intelligent information processing technology for a bioinspired polarization compass is introduced first, which analyzes the research status, development trends, and gap with foreign countries in the field of orientation methods based on atmospheric polarization pattern, as well as the processing methods of the orientation error for a bioinspired polarization compass and integrated system information processing. Subsequently, the noise components of a bioinspired polarization compass and the impact of noise on its directional accuracy is analyzed, introducing the denoising and orientation error compensation technique based on intelligent algorithms such as multi-scale principal component analysis and multi-scale adaptive time-frequency peak filtering. The third part focuses on the application of cubature Kalman filter and their improvement methods in seamless combination orientation systems based on a bioinspired polarization compass. A seamless combination orientation model under discontinuous observation conditions is proposed and a discontinuous observation algorithm based on neural networks is designed.
Chapter1 Introduction1 1.1 Development Background and Research Significance1 1.2 Bioinspired polarization orientation method3 1.3 Orientation error processing method for bioinspired polarization compass13 1.4 Combined orientation system and method for bioinspired polarizaition compass/inertial navigation20 Chapter2 Orientation Method and System for Atmospheric Polarization Pattern27 2.1 Orientation method for atmospheric polarization pattern28 2.1.1 Analysis and automatic identification of neutral point characteristics of atmospheric polarization pattern28 2.1.2 Orientation algorithm based on solar meridian for imaging bioinspired polarization compass32 2.2 Design and integration for bioinspired polarization compass based on FPGA37 2.3 Verification of Bioinspired Polarization compass orientation test41 2.3.1 Static orientation test46 2.3.2 Turntable dynamic orientation test48 2.3.3 UAV airborne dynamic orientation test49 2.4 Chapter Summary53 Chapter3 Processing technology for Bioinspired polarization compass noise55 3.1 Noise analysis for bioinspired polarization compass56 3.1.1 Analysis of the generation mechanism and characteristics for polarization angle image noise56 3.1.2 Analysis of the generation mechanism and characteristics for heading angle data noise63 3.2 Image denoising technology based on multi-scale transformation for bioinspired Polarization compass65 3.2.1 Denoising technology for polarization angle image based on multi-scale transformation68 3.2.2 MS-PCA Image Denoising Technology based on BEMD for Bioinspired Polarization Compass72 3.2.3 Verification of MS-PCA polarization angle image denoising method based on BEMD77 3.3 Heading data denoising technology based on multi-scale transformation for bioinspired polarization compass92 3.3.1 Heading data denoising technology based on multi-scale transformation93 3.3.2 MS-TFPF heading data denoising technology based on EEMD for bioinspired polarization compass96 3.4 Verification of heading data denoising based on multi-scale transformation for bioinspired polarization compass104 3.5 Chapter Summary115 Chapter4 Orientation error modeling and compensation technology for Bioinspired polarization compass118 4.1 Polarization orientation error analysis and model119 4.1.1 Analysis of polarization orientation error119 4.1.2 Model Construction for polarization orientation error125 4.2 Typical neural network models128 4.2.1 Recurrent Neural Networks (RNNs)128 4.2.2 Long Short-Term Memory Neural Networks (LSTMs)133 4.2.3 Gated Recurrent Unit Neural Networks (GRUs)141 4.3 Modeling and compensation of orientation error based on GRU deep learning neural network for bioinspired polarization compass145 4.4 Experimental verification of orientation error model based on GRU deep learning neural network for bioinspired polarization compass152 4.5 Chapter summary156 Chapter5 Seamless combined orientation method and system for bioinspired polarization compass/inertial navigation158 5.1 Seamless combined orientation system for bioinspired polarization compass/inertial navigation160 5.2 Seamless combination orientation model construction for bioinspired polarization compass/inertial navigation162
5.3 Seamless combined orientation method based on self-learning multi-frequency residual correction for bioinspired polarization compass/inertial navigation166 5.4 Experimental verification of the seamless combined orientation method for bioinspired polarization compass/inertial navigation176 5.5 Chapter summary185 Chapter6 Summary and prospect187 6.1 Summary of intelligent information processing technology for bioinspired polarization compass187 6.2 Research outlook190 References192