A scalar-vector combined geomagnetic matching navigation method based on attitude decoupling and heading consistency constraint

Abstract

The application discloses a scalar-vector joint geomagnetic matching navigation method based on attitude decoupling and heading consistency constraint, and belongs to the technical field of geomagnetic matching navigation. The method comprises the following steps: attitude decoupling is carried out on real-time magnetic field data by using a roll angle and a pitch angle, real-time position-related geomagnetic feature vectors and real-time heading-related geomagnetic features are extracted, corresponding geomagnetic feature of candidate points is obtained from a joint geomagnetic map, and position feature residual error is calculated; based on the position feature residual error and inertial navigation prior position, an effective candidate position set is determined, and an inversion heading angle of each candidate position is inverted; a motion consistency constraint cost is constructed, and candidate positions are screened; a joint matching cost is calculated by comprehensively matching residual error, vector matching residual error and motion consistency cost, and inertial navigation information is corrected by selecting an optimal matching position and a heading angle. The application can enhance position constraint by using geomagnetic direction information in real time and dynamically, and improves the precision, robustness and environmental adaptability of geomagnetic matching navigation.

Description

Technical Field

[0001] This invention relates to the field of geomagnetic matching navigation technology, and specifically to a scalar vector joint geomagnetic matching navigation method based on attitude decoupling and heading consistency constraints. Background Technology

[0002] In GNSS-denied environments, geomagnetic matching navigation has become a widely used and effective alternative positioning method. Geomagnetic matching navigation achieves positioning by comparing the data with a pre-built geomagnetic field map, without relying on satellite signals, thus providing stable positioning services even in GNSS-denied environments.

[0003] Traditional geomagnetic matching methods typically rely on either scalar or vector magnetic field information for positioning. Scalar magnetic field data reflects the total field strength, while vector magnetic field data provides more directional information. In some cases, using scalar or vector data alone, or simply fusing them, cannot effectively address mismatches in similar magnetic field regions, leading to decreased matching accuracy. Furthermore, in complex maneuvering environments, existing methods lack the ability to distinguish candidate matching locations, making it difficult to meet navigation requirements under high-dynamic conditions. Traditional methods fail to fully utilize the combined characteristics of scalar and vector magnetic field data, especially during rapid changes in heading, where matching accuracy remains limited, making it difficult to guarantee the stability and accuracy of positioning. Summary of the Invention

[0004] This invention addresses the problems of traditional geomagnetic matching methods, such as easy mismatch in similar magnetic field regions and insufficient ability to distinguish candidate matching positions. It proposes a scalar vector joint geomagnetic matching navigation method based on attitude decoupling and heading consistency constraints.

[0005] The technical solution adopted in this invention is as follows: This invention proposes a scalar vector joint geomagnetic matching navigation method based on attitude decoupling and heading consistency constraints, comprising the following steps: S1. Based on real-time scalar and vector magnetic field measurement data, attitude decoupling is performed using the roll and pitch angles in the carrier attitude information to calculate the real-time position-related geomagnetic feature vector and the real-time heading-related geomagnetic feature vector; the position-related magnetic map feature vector and the heading-related magnetic map feature vector of each candidate point are obtained from the pre-constructed joint geomagnetic map; and the position feature residual of each candidate point is calculated based on the real-time position-related geomagnetic feature vector and the position-related magnetic map feature vector. S2. Based on the position feature residuals and the prior position output by the inertial navigation system, determine the set of valid candidate positions at the current moment; based on the real-time heading-related geomagnetic features and the heading-related magnetic map features of the candidate points, invert the heading angle at each candidate position in the set of valid candidate positions; S3. Obtain the inertial navigation heading and position of adjacent moments output by the inertial navigation system; based on the inverted heading angle, inertial navigation heading and inertial navigation position, construct a motion consistency constraint cost to measure the motion consistency of candidate points between adjacent moments; according to the motion consistency constraint cost, filter the candidate position set at the current moment to obtain the filtered candidate position set and the optimal predecessor relationship corresponding to each candidate point in the set; S4. For each candidate position in the filtered candidate position set, calculate the joint matching cost by combining the scalar magnetic field matching residual, the vector magnetic field matching residual, and the motion consistency constraint cost; select the candidate position with the minimum joint matching cost as the optimal matching position at the current moment, and use its corresponding inversion heading angle as the matching heading angle; use the optimal matching position and the matching heading angle to correct the inertial navigation information output by the inertial navigation system.

[0006] Furthermore, the real-time position-related geomagnetic feature vector is composed of the total magnetic field strength, the horizontal component of the magnetic field, and the vertical component of the magnetic field; the real-time heading-related geomagnetic feature is the azimuth angle of the magnetic field in the horizontal coordinate system after attitude decoupling.

[0007] Furthermore, the position-related magnetic map feature vector is composed of the total magnetic field strength, the horizontal component of the magnetic field, and the vertical component of the magnetic field at the candidate point; the heading-related magnetic map feature is the horizontal projection direction of the geomagnetic field at the candidate point in the navigation coordinate system.

[0008] Furthermore, in S2, the determination of the set of valid candidate positions at the current time includes: The location feature residuals are compared with a preset location feature residual threshold, and candidate points whose location feature residuals are not greater than the location feature residual threshold are selected to form a set of effective candidate locations.

[0009] Further, in S2, the inversion heading angle at each candidate position in the set of inverted valid candidate positions includes: The inverted heading angle is obtained by subtracting the real-time heading-related geomagnetic features from the heading-related magnetic map features of the candidate points and then normalizing the resulting difference.

[0010] Furthermore, the calculation process for the motion consistency constraint cost includes: Get candidate points at the current time inversion heading angle Candidate points at the previous moment inversion heading angle Calculate the candidate heading change ,in, This represents angle normalization operations; Obtain the current and previous inertial navigation directions output by the inertial navigation system, and calculate the change in inertial navigation direction. ; Calculate heading consistency residuals ; Get candidate points at the current time Position coordinates Candidate points at the previous moment Position coordinates Calculate the candidate position increment ; Obtain the current and previous prior positions output by the inertial navigation system, and calculate the inertial navigation position increment. ; Calculate motion continuity residuals ; The motion consistency constraint cost is obtained by weighted summation of the heading consistency residual and the motion continuity residual.

[0011] Furthermore, in S3, the filtering of the candidate position set at the current moment includes: For each candidate position at the current time, search for the candidate position that minimizes the motion consistency constraint cost in the candidate position set of the previous time, and use it as the optimal predecessor candidate position of that candidate position. If the minimum motion consistency constraint cost is not greater than the preset consistency screening threshold, the current candidate position is retained; otherwise, the current candidate position is eliminated.

[0012] Furthermore, the scalar magnetic field matching residual is the absolute value of the difference between the total magnetic field strength measured in real time and the total magnetic field strength at the candidate point. The vector magnetic field matching residual is: ; in, The vector of the magnetic field in the horizontal coordinate system is measured in real time at time k. This is the predicted magnetic field vector obtained by rotating the magnetic field vector of the magnetograph at the candidate point to the horizontal coordinate system based on the inversion heading angle. For vector magnetic field matching residuals; The motion consistency constraint cost is taken as the motion consistency constraint cost under the optimal predecessor relationship corresponding to the candidate point.

[0013] Furthermore, the calculation of the joint matching cost includes: Normalization is performed on the scalar magnetic field matching residual, the vector magnetic field matching residual, and the motion consistency constraint cost, respectively. The joint matching cost is obtained by weighted summation of the three normalized residuals.

[0014] Furthermore, the inertial navigation information output by the inertial navigation system is corrected using the optimal matching position and matching heading angle, including: Let the optimal matching position be The matching heading angle is The prior position output by the inertial navigation system is The prior course is ; Construct the observation residual vector ; in, This represents angle normalization operations; Based on the preset update gain matrix The position and heading of the inertial navigation system are corrected according to the following formula: ; in, and These are the corrected navigation position and heading results, respectively.

[0015] The beneficial effects of this invention are: This invention proposes a scalar-vector joint geomagnetic matching navigation method based on attitude decoupling and heading consistency constraints. Within the prior position range given by inertial navigation, it jointly models scalar geomagnetic features, vector direction features, and heading constraint information at continuous time intervals. Through candidate position heading inversion, heading consistency constraint screening, and scalar-vector joint matching, it achieves the coordinated calculation of the vehicle's position and heading. Compared with traditional geomagnetic matching methods that rely solely on scalar information, solely on vector information, or simply fuse scalar and vector information, this invention improves upon existing methods' problems of prone to mismatching in similar magnetic field regions and insufficient ability to distinguish candidate positions. It also solves the problems of poor matching stability and continuity under complex maneuvering conditions. It can fully utilize the complementary advantages of geomagnetic field strength and direction information, dynamically constrain the temporal evolution process of candidate solutions, improve the accuracy, robustness, and environmental adaptability of geomagnetic matching navigation, and ensure the stability and reliability of navigation and positioning results under complex conditions such as GNSS rejection. Attached Figure Description

[0016] The accompanying drawings, which form part of this application, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0017] Figure 1 This is a flowchart of a scalar vector joint geomagnetic matching navigation method based on attitude decoupling and heading consistency constraints. Detailed Implementation

[0018] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the following embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the scope of protection of the present invention.

[0019] This invention proposes a scalar vector joint geomagnetic matching navigation method based on attitude decoupling and heading consistency constraints, the process of which is as follows: Figure 1 As shown. This method takes a pre-constructed joint geomagnetic map, real-time scalar magnetic field measurement data after magnetic interference compensation, corrected real-time vector magnetic field measurement data, carrier attitude information, and pure inertial navigation calculation results as inputs. Through geomagnetic feature extraction, candidate position and heading inversion, heading consistency constraint screening, and scalar vector joint matching positioning, it achieves joint estimation of position and heading under GNSS rejection conditions.

[0020] In this embodiment, the input quantities, coordinate systems, and symbols involved in the present invention are first uniformly defined.

[0021] Let the first k At each sampling time point, scalar magnetometer measurements and vector magnetometer measurements are acquired. In this embodiment, the measurements refer to the scalar magnetometer measurements after magnetic interference compensation using existing technology. And the vector magnetometer calibrated using existing technology in the carrier coordinate system b The output below, i.e.: Inertial navigation system output roll angle Pitch angle The prior position and prior heading given by a pure inertial navigation system are denoted as follows: in, It is the first k The prior position at each sampling time , It is latitude and longitude. It is the first k The prior heading at each sampling moment.

[0022] The pre-constructed joint geomagnetic map is denoted as: in, Indicates position Scalar geomagnetic characteristics at the location. Indicates position Navigation coordinate system n The vector geomagnetic feature map below.

[0023] The steps of this invention will be described in detail below.

[0024] Step 1: Geomagnetic Feature Extraction In step 1, based on the pre-constructed joint geomagnetic map and real-time magnetic field measurement data, geomagnetic features related to position and heading are extracted by combining the carrier attitude information, providing a feature basis for subsequent candidate position discrimination and candidate heading inversion.

[0025] Since the actual heading of the vehicle at the current moment is yet to be determined, attitude decoupling only uses the roll angle. and pitch angle To preserve the heading difference information required for subsequent heading inversion, no yaw angle prior is introduced. Let the attitude decoupling matrix, composed of roll and pitch angles, be... The real-time vector magnetic field is determined by the carrier coordinate system. b Transform to horizontal coordinate system h The process is represented as: in, For the first k The magnetic field vector in the horizontal coordinate system is measured in real time.

[0026] After the above transformation, the influence of the vehicle's pitch and roll changes on the real-time magnetic field direction observation can be reduced, and the magnetic field vector in the horizontal coordinate system can be obtained for subsequent processing. .

[0027] The location-dependent real-time geomagnetic features are constructed, including the total magnetic field strength, the horizontal component of the magnetic field, and the vertical component of the magnetic field, which are expressed as follows: in, It is the total magnetic field strength, mainly characterizing the magnetic field strength at the current location. and These are the horizontal and vertical components of the magnetic field, respectively, characterizing the distribution of the magnetic field in the horizontal plane and vertical direction after attitude decoupling. Since these three quantities do not explicitly include the vehicle's current heading, they can be used as position-related geomagnetic features for subsequent candidate point selection. Therefore, the... k The real-time location-related geomagnetic feature vector at time t can be represented as: Furthermore, real-time geomagnetic features related to the heading are constructed. Since the horizontal projection direction of the magnetic field after attitude decoupling contains both the absolute direction information of the geomagnetic field and the current orientation information of the carrier, the azimuth angle of the real-time magnetic field in the horizontal coordinate system is defined as: in, This represents the relative azimuth angle of the real-time measured magnetic field in the horizontal coordinate system, which serves as a real-time heading-related geomagnetic feature. This feature will be used for the inversion of the heading corresponding to subsequent candidate positions.

[0028] Based on the current location prior and search radius Construct the first k The set of candidate locations at each sampling time: For any candidate point in the joint geomagnetic map The corresponding magnetic features are read from the scalar geomagnetic feature map and the vector geomagnetic feature map, respectively.

[0029] The scalar magnetogram features at the candidate points are as follows: The vector magnetogram features at the candidate points are: Among them, superscript n Indicates the navigation coordinate system. , and These represent the northward, eastward, and vertical components of the magnetic field at the candidate point, respectively. Correspondingly, the location-related features of the candidate point can be extracted from the magnetic map: Candidate points The location-related magnetograph feature vector is: in, Candidate points The total magnetic field strength at the location of the magnetic field diagram. Candidate points The horizontal component of the magnetic field in the magnetograph. Candidate points The vertical component of the magnetic field in the magnetograph.

[0030] Meanwhile, candidate points The horizontal projection direction of the Earth's magnetic field in the navigation coordinate system can be expressed as: in, This represents the absolute directional characteristics of the geomagnetic field at the candidate point in the navigation coordinate system, serving as the candidate point's heading-related magnetic map feature. This feature will be correlated with the real-time heading angle. They are used together for subsequent candidate course inversion.

[0031] To characterize the degree of matching between real-time measurements and candidate points in location-related geomagnetic features, candidate points are defined. The location feature residual is: in, , , These are the weighting coefficients for the scalar magnetic field matching residual, the horizontal magnetic field residual, and the vertical magnetic field residual, respectively. The smaller the value, the closer the measured magnetic field is to the geomagnetic characteristics of the candidate point, and the greater the likelihood that the candidate point is the current location.

[0032] Therefore, the output of step 1 includes: prior position Prior heading Real-time location-related geomagnetic feature vectors Real-time heading-related geomagnetic features , candidate point location related magnetic map feature vector Candidate point heading-related magnetic map features and location feature residuals The above outputs will serve as inputs for candidate position heading inversion in step 2, as well as for candidate position filtering and joint matching positioning in subsequent steps.

[0033] Step 2: Initial screening of candidate locations and inversion of candidate headings In step 2, within the prior position range given by the inertial navigation, the geomagnetic vector information at the candidate positions is retrieved and combined with the real-time heading-related geomagnetic features extracted in step 1. Invert the candidate headings corresponding to each candidate position.

[0034] Match residuals based on the location features obtained in step 1 , for the k The candidate positions in the candidate position set at each sampling time are initially screened to obtain the effective candidate position set: in, This is the threshold for the location feature residual. For any candidate point... The heading-related magnetic map features of the corresponding candidate points have been obtained from step 1. Meanwhile, the first k Real-time heading-related geomagnetic features If the current carrier is actually located at the candidate point. The difference between the absolute direction of the geomagnetic field at the candidate point in the navigation coordinate system and the relative direction of the magnetic field measured in real time in the vehicle's horizontal coordinate system corresponds to the possible course of the vehicle at that candidate point. Therefore, the candidate point can be obtained through inversion. Corresponding candidate heading angles: in, Assuming the current position is located at a candidate point The inversion heading angle corresponding to the time, This indicates the angle normalization operation.

[0035] Therefore, the output of step 2 is: the set of valid candidate locations. Inversion heading angles corresponding to each candidate point The above results will serve as input for the heading consistency constraint screening in step 3.

[0036] Step 3: Screening of heading consistency and motion continuity constraints In step 3, the candidate positions are narrowed down and filtered by using the relationship between the heading changes and the motion continuity constraints between consecutive moments, and candidate solutions that are inconsistent with the actual motion state of the carrier are eliminated.

[0037] Let the first k The set of valid candidate positions retained after filtering at time -1 is: , No. k The set of valid candidate positions to be filtered at any given time is .when k When =1, the following values ​​are possible: For time First, the change in heading between adjacent time points is calculated using the inertial navigation results: And position increments: For any candidate point in the previous time step and current candidate points The corresponding inversion heading angle is obtained from step 2. and Based on this, candidate heading changes can be constructed: If candidate points are paired If the candidate heading change is consistent with the actual motion of the carrier, then its heading consistency residual should be approximately consistent with the heading change given by the inertial navigation system. Therefore, the heading consistency residual is defined as: Meanwhile, candidate point pairs The position increment on the magnetic map can be expressed as: If the candidate point pair is continuous with the actual trajectory of the carrier, then its position increment should be consistent with the inertial navigation position increment. Therefore, the motion continuity residual is defined as: To comprehensively evaluate the consistency of candidate point pairs in terms of heading changes and position continuity, a motion consistency constraint cost function is constructed: in, and These are the weighting coefficients for the heading consistency residual and the motion continuity residual, respectively.

[0038] For any candidate point at the current moment The candidate set is retained in the previous time step. The search term is the one that most closely matches the candidate predecessor, namely: If the minimum consistency cost is satisfied: Then retain the current candidate point. Otherwise, the candidate point is determined to be inconsistent with the actual motion state of the carrier and is discarded. This yields the... k Candidate set after time-to-time consistency constraint filtering: in, The consistency screening threshold.

[0039] Therefore, step 3 utilizes the consistency between the inversion heading changes and the inertial navigation heading changes between adjacent time points, as well as the continuity between the candidate position increment and the inertial navigation position increment, to recursively shrink the candidate positions, retaining candidate solutions that can form a continuous and reasonable motion trajectory, and eliminating erroneous candidate solutions that, although locally matched at a single time point, do not satisfy the overall motion law. The output of step 3 is the filtered candidate set. and the optimal predecessor relationships corresponding to each candidate point This is used for joint matching and positioning in step 4.

[0040] Step 4: Scalar Vector Joint Matching Localization In step 4, a joint matching cost function is constructed by integrating scalar geomagnetic features, vector direction features, and heading constraint information. This function is used to evaluate each candidate position as a whole and determine the optimal matching position and corresponding heading at the current moment. At the same time, the matching results are used to correct and update the inertial navigation results.

[0041] For any candidate point First, its scalar magnetic field matching residual is defined as: in, For the first k The total magnetic field strength measured in real time at all times. Candidate points The total magnetic field strength at that location. Further, using the inversion heading angle obtained in step 2... Rotate the magnetic field vector at the candidate point to the horizontal coordinate system to obtain the predicted magnetic field vector of the candidate point under the current heading assumption: in, Candidate points The magnetic field vector at that location in the magnetograph. Let be the coordinate rotation matrix determined by the candidate heading angle. To characterize the consistency between the predicted magnetic field direction and the real-time measured magnetic field direction, the vector magnetic field matching residual is defined as: in, For the first k The magnetic field vector in the horizontal coordinate system is measured in real time. The smaller the value, the more consistent the candidate point is with the real-time magnetic field direction under the current heading assumption. Meanwhile, candidate points have already been obtained in step 3. The cost of motion consistency constraints: in, Candidate points The corresponding optimal candidate point number, This represents the cost of motion consistency constraints on the heading change and motion continuity of the candidate point at adjacent time points. To eliminate differences in the dimensions of different residuals, the above three types of residuals are normalized respectively, resulting in: in, , and These are the normalization coefficients for the scalar matching residual, the vector direction residual, and the heading constraint cost, respectively. Based on this, a joint matching cost function is constructed: in, , and These are the weights for scalar features, vector direction features, and heading constraint information, respectively, satisfying that the sum of the three coefficients is 1, and all are greater than or equal to 0. For the candidate set... Calculate the joint matching cost for all candidate points within the range, and select the candidate point with the minimum replacement cost as the optimal matching position at the current moment, i.e.: Therefore, the first k The geomagnetic matching position solution at time t can be expressed as: The corresponding matching heading angle is: To use the matching results to correct and update the inertial navigation results, a position and heading observation residual vector is constructed: Based on this, the inertial navigation results are corrected: in, To update the gain matrix, and These are the corrected navigation position and heading results, respectively. The corrected results will be used for geomagnetic matching navigation at the next moment.

[0042] Therefore, step 4 determines the optimal matching position at the current moment by jointly evaluating the scalar geomagnetic consistency, vector direction consistency, and heading continuity of the candidate positions, and uses the matching results as observation information to correct and update the inertial navigation, thereby improving the positioning accuracy and stability of geomagnetic matching navigation in complex environments.

[0043] In this invention, steps 1 to 4 are executed sequentially, realizing a complete processing flow from raw magnetic field measurement data to candidate position selection, candidate heading inversion, and inertial navigation correction and update. Compared with traditional geomagnetic matching methods that rely solely on scalar information, vector information, or a simple fusion of the two, this invention utilizes position-related and heading-related geomagnetic features extracted by attitude decoupling, and combines them with temporal heading consistency constraints to recursively select candidate solutions. This effectively improves the ability to distinguish candidate positions in similar magnetic field regions and enhances the continuity, stability, and robustness of navigation and positioning under complex maneuvering conditions.

[0044] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the invention. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the invention. Therefore, all technical solutions obtained through equivalent substitution or transformation fall within the protection scope of the present invention.

Claims

1. A scalar vector joint geomagnetic matching navigation method based on attitude decoupling and heading consistency constraints, characterized in that, Includes the following steps: S1. Based on real-time scalar and vector magnetic field measurement data, attitude decoupling is performed using the roll and pitch angles in the carrier attitude information to calculate the real-time position-related geomagnetic feature vector and the real-time heading-related geomagnetic feature. The location-related magnetic map feature vector and heading-related magnetic map feature vector of each candidate point are obtained from the pre-constructed joint geomagnetic map; and the location feature residual of each candidate point is calculated based on the real-time location-related geomagnetic feature vector and the location-related magnetic map feature vector. S2. Based on the position feature residuals and the prior position output by the inertial navigation system, determine the set of valid candidate positions at the current moment; based on the real-time heading-related geomagnetic features and the heading-related magnetic map features of the candidate points, invert the heading angle at each candidate position in the set of valid candidate positions; S3. Obtain the inertial navigation heading and position of adjacent moments output by the inertial navigation system; based on the inverted heading angle, inertial navigation heading and inertial navigation position, construct a motion consistency constraint cost to measure the motion consistency of candidate points between adjacent moments; according to the motion consistency constraint cost, filter the candidate position set at the current moment to obtain the filtered candidate position set and the optimal predecessor relationship corresponding to each candidate point in the set; S4. For each candidate position in the filtered candidate position set, calculate the joint matching cost by combining the scalar magnetic field matching residual, the vector magnetic field matching residual, and the motion consistency constraint cost. The candidate position with the minimum joint matching cost is selected as the optimal matching position at the current moment, and its corresponding inversion heading angle is used as the matching heading angle; the inertial navigation information output by the inertial navigation system is corrected using the optimal matching position and the matching heading angle.

2. The scalar vector joint geomagnetic matching navigation method based on attitude decoupling and heading consistency constraints according to claim 1, characterized in that, The real-time position-related geomagnetic feature vector consists of the total magnetic field strength, the horizontal component of the magnetic field, and the vertical component of the magnetic field; the real-time heading-related geomagnetic feature is the azimuth angle of the magnetic field in the horizontal coordinate system after attitude decoupling.

3. The scalar vector joint geomagnetic matching navigation method based on attitude decoupling and heading consistency constraints according to claim 1, characterized in that, The position-related magnetic map feature vector consists of the total magnetic field strength, the horizontal component of the magnetic field, and the vertical component of the magnetic field at the candidate point; the heading-related magnetic map feature is the horizontal projection direction of the geomagnetic field at the candidate point in the navigation coordinate system.

4. The scalar vector joint geomagnetic matching navigation method based on attitude decoupling and heading consistency constraints according to claim 1, characterized in that, In S2, determining the set of valid candidate positions at the current time includes: The location feature residuals are compared with a preset location feature residual threshold, and candidate points whose location feature residuals are not greater than the location feature residual threshold are selected to form a set of effective candidate locations.

5. The scalar vector joint geomagnetic matching navigation method based on attitude decoupling and heading consistency constraints according to claim 1, characterized in that, In S2, the inversion heading angle at each candidate position in the set of effective candidate positions includes: The inverted heading angle is obtained by subtracting the real-time heading-related geomagnetic features from the heading-related magnetic map features of the candidate points and then normalizing the resulting difference.

6. The scalar vector joint geomagnetic matching navigation method based on attitude decoupling and heading consistency constraints according to claim 1, characterized in that, The calculation process for the motion consistency constraint cost includes: Get candidate points at the current time inversion heading angle Candidate points at the previous moment inversion heading angle Calculate the candidate heading change ,in, This represents angle normalization operations; Obtain the current and previous inertial navigation directions output by the inertial navigation system, and calculate the change in inertial navigation direction. ; Calculate heading consistency residuals ; Get candidate points at the current time Position coordinates Candidate points at the previous moment Position coordinates Calculate the candidate position increment ; Obtain the current and previous prior positions output by the inertial navigation system, and calculate the inertial navigation position increment. ; Calculate motion continuity residuals ; The motion consistency constraint cost is obtained by weighted summation of the heading consistency residual and the motion continuity residual.

7. The scalar vector joint geomagnetic matching navigation method based on attitude decoupling and heading consistency constraints according to claim 1, characterized in that, In S3, the filtering of the candidate position set at the current moment includes: For each candidate position at the current time, search for the candidate position that minimizes the motion consistency constraint cost in the candidate position set of the previous time, and use it as the optimal predecessor candidate position of that candidate position. If the minimum motion consistency constraint cost is not greater than the preset consistency screening threshold, the current candidate position is retained; otherwise, the current candidate position is eliminated.

8. The scalar vector joint geomagnetic matching navigation method based on attitude decoupling and heading consistency constraints according to claim 1, characterized in that, The scalar magnetic field matching residual is the absolute value of the difference between the total magnetic field strength measured in real time and the total magnetic field strength at the candidate point. The vector magnetic field matching residual is: ； in, The vector of the magnetic field in the horizontal coordinate system is measured in real time at time k. This is the predicted magnetic field vector obtained by rotating the magnetic field vector of the magnetograph at the candidate point to the horizontal coordinate system based on the inversion heading angle. For vector magnetic field matching residuals; The motion consistency constraint cost is taken as the motion consistency constraint cost under the optimal predecessor relationship corresponding to the candidate point.

9. The scalar vector joint geomagnetic matching navigation method based on attitude decoupling and heading consistency constraints according to claim 1, characterized in that, The calculation of the joint matching cost includes: Normalization is performed on the scalar magnetic field matching residual, the vector magnetic field matching residual, and the motion consistency constraint cost, respectively. The joint matching cost is obtained by weighted summation of the three normalized residuals.

10. The scalar vector joint geomagnetic matching navigation method based on attitude decoupling and heading consistency constraints according to claim 1, characterized in that, The inertial navigation information output by the inertial navigation system is corrected using the optimal matching position and matching heading angle, including: Let the optimal matching position be The matching heading angle is The prior position output by the inertial navigation system is The prior course is ; Construct the observation residual vector ; in, This represents angle normalization operations; Based on the preset update gain matrix The position and heading of the inertial navigation system are corrected according to the following formula: ； in, and These are the corrected navigation position and heading results, respectively.

Images