A heading angle solving method and system based on sky polarization information

Abstract

This invention relates to the field of polarization navigation technology, and more particularly to a method and system for calculating heading angles based on sky polarization information. The method first acquires polarization distribution data of a sky region, which includes the polarization angle and degree of polarization of multiple sampling points. The polarization distribution data is then filtered based on preset polarization degree filtering conditions to obtain valid data points that meet these conditions. Statistical distribution analysis is performed on these valid data points to determine a target statistical interval that meets the directional consistency condition. Finally, the heading angle is calculated by weighting the polarization degrees of each valid data point within the target statistical interval. This invention effectively suppresses interference from local obstructions, low-polarization-degree regions, and abnormally high polarization noise, reduces dependence on the continuity of the meridian space, and significantly improves the stability, accuracy, and robustness of heading angle calculations in complex environments.

Description

Technical Field

[0001] This invention relates to the field of polarization navigation technology, and in particular to a method and system for calculating heading angle based on sky polarization information. Background Technology

[0002] Sky-based polarization navigation, a crucial technology for autonomous navigation, is widely used in unmanned systems. With the development of optical sensing technology, a complete navigation system has been constructed through the collaborative operation of focal plane sensors, polarization information calculation, and heading estimation. Specifically, the system covers the entire process from light intensity acquisition to angle calculation, including key steps such as polarization angle statistics, threshold selection, and geometric fitting.

[0003] However, existing heading calculation methods directly employ fixed thresholds or geometric fitting without considering the effects of uneven polarization distribution and local occlusion. This can lead to noise interference in low-polarization regions or misleading signals from abnormally high-polarization signals. Specifically, cloud cover or building obstruction can disrupt the integrity of the meridian, causing a decrease in fitting accuracy and thus affecting the stability and accuracy of heading estimation, making it difficult to maintain stable output in complex scenarios. Summary of the Invention

[0004] The present invention aims to at least partially solve one of the technical problems in the related art.

[0005] Therefore, the first objective of this invention is to propose a method for calculating the heading angle based on sky polarization information, comprising the following steps:

[0006] S1, acquire polarization distribution data of the sky region, the polarization distribution data including the polarization angle and polarization degree of multiple sampling points; S2, the polarization distribution data is filtered based on preset polarization degree filtering conditions to obtain valid data points that meet the polarization degree filtering conditions; S3, Perform statistical distribution analysis on the effective data points to determine the target statistical interval that meets the directional consistency condition; S4. The heading angle is calculated by weighting the polarization degree of each valid data point within the target statistical interval.

[0007] In one embodiment of the present invention, S1 further includes: S11, use a focal plane polarization sensor to synchronously collect light intensity information in different polarization directions, and use the light intensity information for pixel-level polarization calculation to generate initial polarization data; S12, calculate the polarization angle and polarization degree of each pixel based on the initial polarization data to obtain polarization distribution data containing the polarization angle and polarization degree of multiple sampling points.

[0008] In one embodiment of the present invention, S2 further includes: S21, Set a polarization degree threshold, and use the polarization degree threshold to traverse each sampling point in the polarization distribution data; S22, retain pixels with polarization degrees greater than the polarization degree threshold as valid data points to suppress noise and blurred edge interference generated in low polarization degree regions.

[0009] In one embodiment of the present invention, S3 further includes: S31, based on the angle-polarization degree two-dimensional binning method, divides the selected meridion candidate pixels into multiple two-dimensional statistical intervals according to the polarization angle and polarization degree; S32 counts the number of data points within each two-dimensional statistical interval to obtain frequency statistics reflecting the joint distribution of polarization angle and polarization degree.

[0010] In one embodiment of the present invention, S32 further includes: S321, Select a two-dimensional interval with a set number of frequencies, and limit the polarization angle ranges corresponding to the two-dimensional intervals to not exceed a set range; S322, the two-dimensional interval that meets the phase difference limit of the polarization angle range is determined as the target statistical interval that meets the direction consistency condition, so as to eliminate abnormal angles caused by random noise.

[0011] In one embodiment of the present invention, S3 further includes: S31, based on the kernel density estimation method, divides the selected meridional candidate pixels into multiple two-dimensional statistical intervals according to the polarization angle and polarization degree; S32 counts the number of data points within each two-dimensional statistical interval to obtain frequency statistics reflecting the joint distribution of polarization angle and polarization degree.

[0012] In one embodiment of the present invention, S4 further includes: S41, the selected interval's angle and polarization range form a joint statistical region, and the joint statistical region is used to calculate the heading angle estimate; S42, within the joint statistical region, a weighted calculation is performed based on the polarization degree and quantity distribution of each pixel to finally obtain the estimated heading angle.

[0013] In one embodiment of the present invention, the calculation formula for the kernel density estimation method is as follows:

[0014] in, It is the polarization angle. For degree of polarization, K (·) is a two-dimensional kernel function. This refers to the bandwidth parameter in the angular direction. This represents the bandwidth parameter in the polarization direction. To set weight values.

[0015] To achieve the above objectives, a second aspect of the present invention proposes a heading angle calculation system based on sky polarization information, comprising: The data acquisition module acquires polarization distribution data of the sky region, the polarization distribution data including the polarization angle and degree of polarization of multiple sampling points; The data point filtering module filters the polarization distribution data based on preset polarization degree filtering conditions to obtain valid data points that meet the polarization degree filtering conditions. The interval division module performs statistical distribution analysis on the effective data points to determine the target statistical interval that meets the directional consistency condition; The heading angle calculation module performs weighted processing based on the polarization degree of each valid data point within the target statistical interval to calculate the heading angle.

[0016] To achieve the above objectives, a third aspect of the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in the first aspect.

[0017] The method, system, and storage medium of this invention, through polarization degree weighting and statistical filtering mechanisms, effectively suppress local occlusion and noise interference, and significantly improve the stability, accuracy, and robustness of heading angle calculation in complex environments.

[0018] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0019] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a flowchart of a heading angle calculation method based on sky polarization information according to an embodiment of the present invention; Figure 2 This is a structural diagram of a heading angle calculation system based on sky polarization information according to an embodiment of the present invention. Detailed Implementation

[0020] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0021] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0022] The following describes, with reference to the accompanying drawings, a method and system for calculating heading angle based on sky polarization information according to an embodiment of the present invention.

[0023] Example 1 Figure 1 This is a flowchart of a heading angle calculation method based on sky polarization information according to an embodiment of the present invention.

[0024] like Figure 1 As shown, the heading angle calculation method based on sky polarization information includes the following steps: S1, acquire polarization distribution data of the sky region, the polarization distribution data includes the polarization angle and degree of polarization of multiple sampling points.

[0025] Further, step S1 includes: S11, use a focal plane polarization sensor to synchronously collect light intensity information in different polarization directions, and use the light intensity information for pixel-level polarization calculation to generate initial polarization data.

[0026] Specifically, optical sensor devices are used to probe the sky region to collect optical signals that reflect the sky's polarization patterns. Each optical sensor corresponds to a sampling point, and each sampling point corresponds to a spatial location in the sky region. The optical signal corresponding to this spatial location includes polarization data for that location, which represents the light intensity information in different polarization directions.

[0027] S12, calculate the polarization angle and polarization degree of each pixel based on the initial polarization data to obtain polarization distribution data containing the polarization angle and polarization degree of multiple sampling points.

[0028] By acquiring complete polarization distribution data, we can comprehensively reflect the spatial distribution characteristics of the sky polarization field, providing a sufficient data foundation for subsequent removal of low-quality data and extraction of robust meridian features, and ensuring that the heading angle calculation process has sufficient information redundancy.

[0029] This step establishes a digital representation of sky polarization information by acquiring distribution data including polarization angle and degree of polarization, enabling subsequent processing to be based on multi-dimensional polarization characteristics. Comprehensive data acquisition ensures that no effective polarization signals are missed, laying a data foundation for improving the accuracy and robustness of heading angle calculation.

[0030] S2, the polarization distribution data is filtered based on preset polarization degree filtering conditions to obtain valid data points that meet the polarization degree filtering conditions.

[0031] Furthermore, step S2 includes: S21, Set a polarization degree threshold, and use the polarization degree threshold to traverse each sampling point in the polarization distribution data.

[0032] Specifically, the filtering criteria are set based on the noise distribution characteristics and signal strength requirements in the actual application scenario. These criteria can be fixed numerical limits or dynamically adjusted ranges. The core principle is to use the polarization degree amplitude to distinguish between effective signals and background noise. As one implementation method, in existing technologies, a polarization degree threshold is often set to a value greater than 0.3.

[0033] S22, retain pixels with polarization degrees greater than the polarization degree threshold as valid data points to suppress noise and blurred edge interference generated in low polarization degree regions.

[0034] The above-mentioned filtering mechanism can effectively suppress invalid polarization information, reduce the interference of local occlusion and low polarization region on heading angle calculation, and improve the stability and accuracy of subsequent statistical analysis and heading angle calculation.

[0035] S3, Perform statistical distribution analysis on the effective data points to determine the target statistical interval that meets the directional consistency condition.

[0036] Further, step S3 includes: S31 divides the selected meridian candidate pixels into multiple two-dimensional statistical intervals according to the polarization angle and polarization degree.

[0037] One implementation method is to use an angle-polarization degree two-dimensional binning statistical method to partition the pixels. Specifically, the meridion candidate pixels are divided into multiple two-dimensional planning intervals according to the polarization angle and polarization degree. The step size of the polarization angle is 0.1°, and the step size of the polarization degree is 0.1.

[0038] As another implementation method, a probabilistic statistical method based on kernel density estimation (KDE) can be used to partition the pixels. Compared with the angle-polarization degree two-dimensional binning statistical method, the probabilistic statistical method based on kernel density estimation is a continuous probability density estimation, which can obtain a smoother and more robust principal direction determination.

[0039] Specifically, the formula for calculating kernel density estimation is as follows:

[0040] in, It is the polarization angle. For degree of polarization, K (·) is a two-dimensional kernel function. This refers to the bandwidth parameter in the angular direction. This represents the bandwidth parameter in the polarization direction. To set the weight value, this weight value is usually taken as the pixel polarization degree or a function thereof (e.g., ...). =Pi or Pi 2 This method enhances high-reliability polarization information. By performing an extremum search on the probability density function, the principal direction region corresponding to the density peak is obtained, which is the most likely direction of the solar meridian. Simultaneously, a probability threshold region can be set near the density peak, and samples within this region are weighted and averaged to obtain the final heading angle estimate. Compared to traditional binning statistical methods, the kernel density estimation-based scheme does not require fixed bin boundaries, avoiding quantization errors and boundary effects, resulting in a more continuous and smooth angle distribution. It also exhibits stronger robustness to local occlusion, outlier noise, and abnormally high polarization points, thereby further improving the stability and accuracy of the heading angle calculation.

[0041] S32 counts the number of data points within each two-dimensional statistical interval to obtain frequency statistics reflecting the joint distribution of polarization angle and polarization degree.

[0042] S321, Select a two-dimensional interval with a set number of frequencies, and limit the polarization angle ranges corresponding to the two-dimensional intervals to not exceed a set range.

[0043] As one implementation method, selecting two two-dimensional intervals results in the intervals being distributed along a straight line, making it impossible to form a closed range. Selecting too many intervals, on the other hand, can lead to interval dispersion, causing the intervals to be distributed in the low-polarization region. Therefore, in practical applications, three two-dimensional intervals are often chosen. S322, the two-dimensional interval that meets the phase difference limit of the polarization angle range is determined as the target statistical interval that meets the direction consistency condition, so as to eliminate abnormal angles caused by random noise.

[0044] This step replaces traditional geometric fitting with statistical distribution analysis, reducing the reliance on the continuity of the data space. Combined with the directional consistency condition, it can effectively eliminate outlier noise with dispersed angles, ensuring that the target statistical interval reflects the true characteristics of the solar meridian, thereby significantly improving the robustness and stability of the heading angle calculation in complex environments.

[0045] S4. The heading angle is calculated by weighting the polarization degree of each valid data point within the target statistical interval.

[0046] Further, step S4 includes: S41, the angle and polarization range of the three selected intervals are used to form a joint statistical region, and the joint statistical region is used to calculate the heading angle estimate; S42, within the joint statistical region, a weighted calculation is performed based on the polarization degree and quantity distribution of each pixel to finally obtain the estimated heading angle.

[0047] In practice, a weighted calculation can be performed based on the polarization degree values ​​and their distribution quantity of data points within the region, ultimately outputting an estimated heading angle. This step, by introducing a polarization degree weighting mechanism, ensures that high-reliability polarization information dominates the heading calculation, thereby significantly improving the accuracy and robustness of the heading angle calculation results. It effectively suppresses the interference of low polarization degree noise and outliers on the final output results, ensuring the stability of heading estimation in complex environments.

[0048] Example 2 To further illustrate the specific execution process of the method of the present invention in practical applications, sky polarization data collected in a real-world environment was selected for processing and verification. During the experiment, a split-plane polarization camera was used to continuously observe the sky region, collecting a total of 100 sets of polarization image data. Each set of data contained pixel-level polarization angle and polarization degree information. After the data acquisition was completed, calculations were performed sequentially according to the processing flow of the method of the present invention.

[0049] First, pixel-level calculations are performed on each set of original polarization images to obtain complete polarization distribution data, i.e., the polarization angle and degree of polarization information corresponding to each pixel. Then, this distribution data is processed through a series of steps to enter the data filtering stage. A polarization degree threshold of 0.3 is set, and each pixel is evaluated individually. Data points with a polarization degree greater than 0.3 are retained as valid data, while the remaining data points are discarded, thus completing the transformation from raw data to a valid dataset.

[0050] After obtaining valid data points, statistical distribution analysis is performed. Specifically, all valid data points are divided into two dimensions according to polarization angle and polarization degree, with a polarization angle division step of 0.1° and a polarization degree division step of 0.1°. The number of data points in each interval is counted to form a two-dimensional frequency distribution. Then, all statistical intervals are sorted by frequency, and the three intervals with the highest frequency are selected. It is further determined whether the polarization angle ranges corresponding to these three intervals meet the preset angle consistency condition (e.g., deviation not exceeding 2°). If the condition is met, these three intervals are jointly determined as the target statistical interval; if not, the intervals are reselected and the judgment process is repeated.

[0051] After determining the target statistical interval, all data points within that interval are weighted. Specifically, the polarization degree of each pixel is used as the weight, and a weighted average of its corresponding polarization angles is calculated to obtain the estimated heading angle for the current data set. This process is repeated for each of the 100 data sets to complete the entire heading angle calculation.

[0052] As can be seen from the above processing flow, the method of this invention forms a complete and clear processing chain from raw data acquisition, data filtering, statistical analysis to final heading angle calculation. The steps are closely linked, and the data converges step by step: from the original distributed data to the effective data, then to the target statistical interval, and finally obtaining a stable heading angle estimation result. This process avoids noise interference caused by direct calculation based on global data, and improves the stability and consistency of the solution process through step-by-step filtering and statistical constraints.

[0053] In actual processing results, all sets of data were able to stably complete the above process, and the heading angle output was continuous without obvious jumps, indicating that the method has good feasibility and stability in engineering applications and can meet the requirements of real-time performance and robustness of actual systems.

[0054] Example 3 A heading angle calculation system 10 based on sky polarization information includes: The data acquisition module 100 acquires polarization distribution data of the sky region, the polarization distribution data including the polarization angle and degree of polarization of multiple sampling points; The data point filtering module 200 filters the polarization distribution data based on preset polarization degree filtering conditions to obtain valid data points that meet the polarization degree filtering conditions. The interval division module 300 performs statistical distribution analysis on the effective data points to determine the target statistical interval that meets the directional consistency condition. The heading angle calculation module 400 performs weighted processing based on the polarization degree of each valid data point within the target statistical interval to calculate the heading angle.

[0055] When performing statistical distribution analysis on the effective data points, the interval division module 300 can use either the angle-polarization degree two-dimensional binning method or the kernel density estimation method.

[0056] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described method for calculating heading angle based on sky polarization information.

[0057] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0058] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

Claims

1. A method for calculating heading angle based on sky polarization information, characterized in that, Includes the following steps: S1, acquire polarization distribution data of the sky region, the polarization distribution data including the polarization angle and polarization degree of multiple sampling points; S2, the polarization distribution data is filtered based on preset polarization degree filtering conditions to obtain valid data points that meet the polarization degree filtering conditions; S3, Perform statistical distribution analysis on the effective data points to determine the target statistical interval that meets the directional consistency condition; S4. The heading angle is calculated by weighting the polarization degree of each valid data point within the target statistical interval.

2. The method as described in claim 1, characterized in that, S1 further includes: S11, use a focal plane polarization sensor to synchronously collect light intensity information in different polarization directions, and use the light intensity information for pixel-level polarization calculation to generate initial polarization data; S12, calculate the polarization angle and polarization degree of each pixel based on the initial polarization data to obtain polarization distribution data containing the polarization angle and polarization degree of multiple sampling points.

3. The method as described in claim 1, characterized in that, S2 further includes: S21, Set a polarization degree threshold, and use the polarization degree threshold to traverse each sampling point in the polarization distribution data; S22, retain pixels with polarization degrees greater than the polarization degree threshold as valid data points to suppress noise and blurred edge interference generated in low polarization degree regions.

4. The method as described in claim 1, characterized in that, S3 further includes: S31, based on the angle-polarization degree two-dimensional binning method, divides the selected meridional candidate pixels into multiple two-dimensional statistical intervals according to the polarization angle and polarization degree; S32 counts the number of data points within each two-dimensional statistical interval to obtain frequency statistics reflecting the joint distribution of polarization angle and polarization degree.

5. The method as described in claim 4, characterized in that, S32 further includes: S321, Select a two-dimensional interval with a set number of frequencies, and limit the polarization angle ranges corresponding to the two-dimensional intervals to not exceed a set range; S322, the two-dimensional interval that meets the phase difference limit of the polarization angle range is determined as the target statistical interval that meets the direction consistency condition, so as to eliminate abnormal angles caused by random noise.

6. The method as described in claim 1, characterized in that, S3 further includes: S31, based on the kernel density estimation method, divides the selected meridional candidate pixels into multiple two-dimensional statistical intervals according to the polarization angle and polarization degree; S32 counts the number of data points within each two-dimensional statistical interval to obtain frequency statistics reflecting the joint distribution of polarization angle and polarization degree.

7. The method as described in claim 1, characterized in that, S4 further includes: S41, the selected interval's angle and polarization range form a joint statistical region, and the joint statistical region is used to calculate the heading angle estimate; S42, within the joint statistical region, a weighted calculation is performed based on the polarization degree and quantity distribution of each pixel to finally obtain the estimated heading angle.

8. The method as described in claim 1, characterized in that, The calculation formula for the kernel density estimation method is as follows: in, It is the polarization angle. For degree of polarization, K (·) is a two-dimensional kernel function. This refers to the bandwidth parameter in the angular direction. This represents the bandwidth parameter in the polarization direction. To set weight values.

9. A heading angle calculation system based on sky polarization information, characterized in that, include: The data acquisition module acquires polarization distribution data of the sky region, the polarization distribution data including the polarization angle and degree of polarization of multiple sampling points; The data point filtering module filters the polarization distribution data based on preset polarization degree filtering conditions to obtain valid data points that meet the polarization degree filtering conditions. The interval division module performs statistical distribution analysis on the effective data points to determine the target statistical interval that meets the directional consistency condition; The heading angle calculation module performs weighted processing based on the polarization degree of each valid data point within the target statistical interval to calculate the heading angle.

10. A computer-readable storage medium storing a computer program that, when executed by a processor, implements the method of any one of claims 1-8.

Images