Indoor local space equivalent magnetic dipole model parameter inversion and positioning method

Abstract

The application relates to the technical field of indoor positioning, and provides an indoor local space equivalent magnetic dipole model parameter inversion and positioning method. The method simultaneously uses magnetic field data collected at continuous multiple time points for modeling, and explicitly introduces carrier pose information as a to-be-estimated parameter in the magnetic field model; by constructing a magnetic field observation equation containing the pose variable, the equivalent magnetic dipole model parameters and the pose transformation of the carrier at each time point are jointly solved by using the magnetic field measurement values at continuous time points, so that the deep fusion of modeling and positioning is realized. The method breaks through the limitations of traditional methods which depend on single-time modeling and coverage range assumptions, effectively suppresses the accumulation of model prediction errors, and improves the accuracy and continuity of pose estimation in an indoor complex magnetic field environment.

Description

Technical Field

[0001] This application relates to the field of indoor positioning technology, and in particular to a method for parameter inversion and positioning of an equivalent magnetic dipole model in indoor local space. Background Technology

[0002] As a globally distributed natural vector field, the Earth's magnetic field exhibits excellent spatiotemporal stability and spatial homogeneity in open outdoor environments. Based on this physical property, geomagnetic information has long served as a heading reference in navigation systems, compensating for heading drift caused by the accumulation of gyroscope bias and random walk noise over time in inertial navigation systems. By fusing the three-axis output of a magnetometer with gyroscope data, geomagnetic-assisted heading correction technology can effectively improve the orientation accuracy of inertial navigation systems. Related research shows that this technology can reduce azimuth error by approximately 60% to 80%.

[0003] As navigation scenarios shift from outdoors to indoors, the magnetic field environment changes significantly. Influenced by factors such as the shielding effect of reinforced concrete structures on the Earth's magnetic field, alternating electromagnetic interference generated by electrical equipment, and local field distortions caused by metal components, the indoor magnetic field exhibits a complex, spatially non-uniform distribution. This distortion characteristic renders traditional heading calculation models based on the uniform field assumption ineffective, thus limiting the application of existing magnetic navigation technology in indoor environments.

[0004] It is noteworthy that, despite the highly non-uniform distribution of indoor magnetic fields, their spatial structure still possesses certain regularities and identifiable characteristics. Measured data shows that the magnetic field vector corresponding to a specific indoor location remains relatively stable over time and exhibits a significant correlation with spatial coordinates. This spatial distribution characteristic of the magnetic field, modulated by both building structure and the electromagnetic environment—the so-called "magnetic fingerprint"—makes it a potential source of positioning information in scenarios where global navigation satellite system signals are difficult to cover, such as inside buildings, underground spaces, and underwater areas. In recent years, indoor positioning technology based on magnetic field spatial characteristics has gradually become a hot research area in navigation under restricted environments.

[0005] Indoor positioning technologies based on magnetic field spatial characteristics mainly develop along two directions: one is the absolute positioning method based on magnetic field fingerprint matching, which achieves positioning by matching real-time observations with a fingerprint database in the online stage after pre-constructing a magnetic field fingerprint map; the other is the relative pose estimation method based on magnetic field sequence correlation, which uses magnetic field data continuously observed during the vehicle's movement to recursively deduce the vehicle's relative displacement by the correlation between magnetic field observations at adjacent times, i.e., "magnetic odometry". These two technical routes represent two typical paradigms in the field of indoor magnetic field positioning: the former emphasizes the unique representation of spatial location and relies on the construction of high-resolution prior maps; the latter focuses on the continuity constraints of the movement process and emphasizes the correlation mining of sequence data.

[0006] However, both methods suffer from the inherent limitation of separating modeling and localization. Fingerprint matching treats magnetic field modeling as an offline pre-construction step, requiring the prior acquisition of a high-precision magnetic field fingerprint database. Its localization accuracy is highly dependent on the density and timeliness of the fingerprint database. On the one hand, the construction and maintenance of the fingerprint database requires a significant investment of time and manpower, and data needs to be re-acquired when the indoor environment layout changes or electromagnetic interference sources change, limiting its applicability. On the other hand, it is difficult to adapt to dynamic environmental changes, and its localization performance significantly degrades when environmental characteristics change.

[0007] Magnetic field odometry (FMO) technology enables online recursive estimation of pose, but its performance is highly dependent on the accuracy of magnetic field modeling and the precision of correlation between magnetic field models at adjacent time points. Existing FMO methods typically perform local modeling based on magnetic field data acquired at a single time point, implicitly assuming that the effective coverage of the constructed magnetic field model can encompass the magnetic field sampling area at the next time point. However, this assumption is difficult to strictly hold during actual vehicle movement: when the vehicle moves at high speeds or the magnetic field space changes drastically, a deviation may occur between the model coverage and the actual sampling location at the next time point, causing the prediction of the magnetic field at the next time point based on the current time-based model to fail. Furthermore, even if the coverage meets the assumption, the predicted value of the magnetic field model at the sampling point at the next time point, calculated solely from the current time-based data, will still inevitably differ from the actually acquired magnetic field value. This difference accumulates gradually with movement, reducing the accuracy and robustness of pose estimation. Summary of the Invention

[0008] In view of this, embodiments of this application provide a method for parameter inversion and positioning of an equivalent magnetic dipole model in an indoor local space, in order to solve the problem of limited positioning accuracy and continuity of existing magnetic navigation technology in indoor local spaces.

[0009] A first aspect of this application provides a method for inverting and locating parameters of an equivalent magnetic dipole model for an indoor local space, comprising:

[0010] An array of magnetic sensors is installed in a localized indoor space to acquire data from the array of magnetic sensors. Spatial magnetic field measurement value at time and The spatial magnetic field measurement at that moment;

[0011] An equivalent magnetic dipole model and a magnetic field observation model are constructed. The equivalent magnetic dipole model is used to predict the space magnetic field. The magnetic field observation model is used to characterize the relationship between the measured values ​​and the predicted values ​​of the space magnetic field.

[0012] Based on the constructed equivalent magnetic dipole model and magnetic field observation model, using Time and Spatial magnetic field measurements at time t and predictions from the equivalent magnetic dipole model Time and Predicted spatial magnetic field values ​​at time t, and calculation of equivalent magnetic dipole model parameters. Time's up The carrier pose transformation parameters at any given time;

[0013] The carrier is positioned based on the parameters of the equivalent magnetic dipole model and the carrier pose transformation parameters.

[0014] A second aspect of this application provides a device for inverting and locating parameters of an equivalent magnetic dipole model for an indoor local space, comprising:

[0015] The measurement module is configured to place an array of magnetic sensors in a localized indoor space to acquire data from the array of magnetic sensors. Spatial magnetic field measurement value at time and The spatial magnetic field measurement at that moment;

[0016] The building module is configured to build an equivalent magnetic dipole model and a magnetic field observation model; the equivalent magnetic dipole model is used to predict the space magnetic field; the magnetic field observation model is used to characterize the relationship between the measured values ​​and the predicted values ​​of the space magnetic field.

[0017] The inversion module, based on the constructed equivalent magnetic dipole model and magnetic field observation model, utilizes... Time and Spatial magnetic field measurements at time t and predictions from the equivalent magnetic dipole model Time and Predicted spatial magnetic field values ​​at time t, and calculation of equivalent magnetic dipole model parameters. Time's up The carrier pose transformation parameters at any given time;

[0018] The positioning module is configured to position the carrier based on the parameters of the equivalent magnetic dipole model and the carrier pose transformation parameters.

[0019] A third aspect of this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described method.

[0020] A fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described method.

[0021] The beneficial effects of this application embodiment compared with the prior art are as follows: This application embodiment uses magnetic field data collected at multiple consecutive moments for modeling, and explicitly introduces the carrier pose information as the parameter to be estimated into the magnetic field model. By constructing a magnetic field observation equation that includes pose variables, the equivalent magnetic dipole model parameters and the pose transformation of the carrier at each moment are jointly solved using the magnetic field measurement values ​​at consecutive moments, thereby achieving deep integration of modeling and positioning. The technical solution provided by this application embodiment breaks through the limitations of traditional methods that rely on single-moment modeling and coverage assumptions, effectively suppresses the accumulation of model prediction errors, and improves the accuracy and continuity of pose estimation in complex indoor magnetic field environments. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a flowchart illustrating a method for inverting and locating parameters of an equivalent magnetic dipole model for an indoor local space, as provided in an embodiment of this application.

[0024] Figure 2 This is a schematic diagram of the distribution and coverage of magnetic dipoles in an equivalent magnetic dipole model based on Fibonacci space sampling.

[0025] Figure 3 yes Time's up A schematic diagram of the pose transformation relationship of the carrier at any given time.

[0026] Figure 4 This is a flowchart illustrating another method for inverting and locating parameters of an equivalent magnetic dipole model for indoor local space provided in this application embodiment.

[0027] Figure 5 This is a schematic diagram of an indoor local space equivalent magnetic dipole model parameter inversion and positioning device provided in an embodiment of this application.

[0028] Figure 6 This is a schematic diagram of the electronic device provided in the embodiments of this application. Detailed Implementation

[0029] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0030] The following will describe in detail, with reference to the accompanying drawings, a method and apparatus for inverting and locating parameters of an equivalent magnetic dipole model for an indoor local space according to an embodiment of this application.

[0031] As mentioned above, fingerprint matching treats magnetic field modeling as an offline pre-construction step, requiring the prior acquisition of a high-precision magnetic field fingerprint database. Its positioning accuracy is highly dependent on the density and timeliness of the fingerprint database. Simultaneously, the performance of magnetic field odometry technology is highly dependent on the accuracy of magnetic field modeling and the precision of the correlation between magnetic field models at adjacent time points.

[0032] Existing magnetic field modeling methods mainly include Gaussian process models, magnetic field polynomial models, and equivalent magnetic dipole models.

[0033] The Gaussian process model, based on the irrotational and divergent physical properties of the magnetic field, decomposes the magnetic field into mathematical expressions of scalar or vector potentials by designing a covariance kernel function. Differential constraints are introduced to jointly model the three components, ensuring the model's global physical consistency. During modeling, the Gaussian process model uses magnetic field data collected at the current moment to optimize the model's hyperparameters online. Hilbert space approximation techniques (such as Fourier feature expansion) are used to transform the covariance matrix into a low-rank basis function combination to reduce real-time computational complexity. The model's advantage lies in its ability to explicitly incorporate physical laws; however, its limitations include a high dependence on prior knowledge in kernel function design, and a significant increase in extrapolation errors in regions far from the observation point when relying only on data within the limited field of view at the current moment. Furthermore, traditional isotropic kernels struggle to capture the directionality and heterogeneity of the real magnetic field (such as gradient changes caused by building structures), making it difficult to accurately describe complex local magnetic field details based on single-moment data.

[0034] The magnetic field polynomial model directly fits the spatial distribution of magnetic field intensity within the observation region at the current moment using polynomial functions. Structurally, it explicitly embeds rotation-free and divergence-free differential constraint equations, thus avoiding the computational burden caused by complex kernel functions and covariance matrix operations in Gaussian process models. In real-time applications, the magnetic field polynomial model rapidly solves for polynomial coefficients using magnetic field sampling data at the current moment, achieving instantaneous modeling of the local magnetic field. Its advantages lie in high computational efficiency and strong real-time performance, but its drawbacks include the limited expressive power of polynomial basis functions, making it difficult to describe abrupt changes in the local magnetic field or high-dimensional nonlinear relationships. Furthermore, the extrapolation performance of polynomial models built based on single-moment data is poor, limiting its applicability to local regions near the observation point.

[0035] The equivalent magnetic dipole model constructs a virtual equivalent magnetic dipole, ensuring that the magnetic field it generates within the observation region at the current moment is consistent with the actual sampled data. Magnetic dipole parameters (such as magnetic moment and position) are inverted using the magnetic field data acquired at the current moment to construct an equivalent magnetic source that reflects the physical characteristics of the magnetic field source. This model is simple in structure, computationally efficient, and supports rapid parameter inversion based on single-moment observation data. By superimposing multiple magnetic dipoles, this model can approximate complex local magnetic field distributions using multi-point sampled data at the current moment, achieving high-precision magnetic field prediction. Compared to the previous two models, the equivalent magnetic dipole model combines physical interpretability with computational efficiency, making it more suitable for online applications in magnetic field odometry involving magnetic field modeling and pose recursion.

[0036] However, the aforementioned models are mostly based on local modeling using magnetic field data collected at a single moment, and implicitly assume that the effective coverage of the constructed magnetic field model can include the magnetic field sampling area at the next moment. In actual carrier motion, both the robustness and prediction accuracy of the model will be affected.

[0037] In view of this, embodiments of this application provide a method for inverting and locating parameters of an equivalent magnetic dipole model in an indoor local space. This method uses magnetic field data collected at multiple consecutive moments for modeling, and explicitly introduces the carrier pose information as a parameter to be estimated into the magnetic field model. By constructing a magnetic field observation equation that includes pose variables, the equivalent magnetic dipole model parameters and the carrier pose transformation at each moment are jointly solved using magnetic field measurements at consecutive moments, thereby achieving deep integration of modeling and location. This method overcomes the limitations of traditional methods that rely on single-moment modeling and coverage assumptions, effectively suppresses the accumulation of model prediction errors, and improves the accuracy and continuity of pose estimation in complex indoor magnetic field environments.

[0038] Figure 1 This is a flowchart illustrating a method for parameter inversion and localization of an equivalent magnetic dipole model in an indoor local space, as provided in an embodiment of this application. Figure 1 As shown, the method includes the following steps:

[0039] In step S101, an array of magnetic sensors is set in a localized indoor space to acquire data from the array of magnetic sensors. Spatial magnetic field measurement value at time and Measurement of the spatial magnetic field at a given time.

[0040] In step S102, an equivalent magnetic dipole model and a magnetic field observation model are constructed.

[0041] Among them, the equivalent magnetic dipole model is used to predict the space magnetic field; the magnetic field observation model is used to characterize the relationship between the measured value and the predicted value of the space magnetic field.

[0042] In step S103, based on the constructed equivalent magnetic dipole model and magnetic field observation model, using... Time and Spatial magnetic field measurements at time t and predictions from the equivalent magnetic dipole model Time and Predicted spatial magnetic field values ​​at time t, and calculation of equivalent magnetic dipole model parameters. Time's up The carrier pose transformation parameters at any given time.

[0043] In step S104, the carrier is positioned based on the equivalent magnetic dipole model parameters and the carrier pose transformation parameters.

[0044] In some embodiments of this application, the method may be executed by a server or by a terminal device with certain processing capabilities.

[0045] In some embodiments of this application, an array of magnetic sensors can be first installed in a localized indoor space, and the data from the array of magnetic sensors can be acquired. Spatial magnetic field measurement value at time and Measurement of the spatial magnetic field at a given time.

[0046] The magnetic sensor can be a triaxial magnetometer or other magnetic sensors; there are no restrictions here. The number of magnetic sensors in the array magnetic sensor can also be set according to actual needs; there are no restrictions here.

[0047] In some embodiments of this application, an equivalent magnetic dipole model for predicting space magnetic fields and a magnetic field observation model for characterizing the relationship between measured and predicted space magnetic fields can be constructed.

[0048] Furthermore, based on the constructed equivalent magnetic dipole model and magnetic field observation model, it is possible to utilize... Time and Spatial magnetic field measurements at time t and predictions from the equivalent magnetic dipole model Time and Predicted spatial magnetic field values ​​at time t, and calculation of equivalent magnetic dipole model parameters. Time's up The carrier pose transformation parameters at any given time.

[0049] Finally, the carrier can be located based on the parameters of the equivalent magnetic dipole model and the carrier pose transformation parameters. The localization algorithm can be selected according to actual needs, and no restrictions are imposed here.

[0050] According to the technical solution provided in the embodiments of this application, magnetic field data collected at multiple consecutive moments are used simultaneously for modeling, and the carrier pose information is explicitly introduced into the magnetic field model as the parameter to be estimated. By constructing a magnetic field observation equation that includes pose variables, the equivalent magnetic dipole model parameters and the pose transformation of the carrier at each moment are jointly solved using the magnetic field measurement values ​​at consecutive moments, thereby achieving deep integration of modeling and positioning. The technical solution provided in the embodiments of this application breaks through the limitations of traditional methods that rely on single-moment modeling and coverage assumptions, effectively suppresses the accumulation of model prediction errors, and improves the accuracy and continuity of pose estimation in complex indoor magnetic field environments.

[0051] In some embodiments of this application, the magnetic dipole position of the equivalent magnetic dipole model can be determined in the following manner:

[0052] Determine the golden ratio of each magnetic dipole: ;

[0053] Determine the golden angle of each magnetic dipole: ; This is the magnetic dipole number;

[0054] Will Magnetic dipoles are distributed around a circle centered at coordinate point (0,0,0). On a sphere with radius r, we obtain the first... A magnetic dipole Axis coordinates : ; It is a positive integer;

[0055] Determine the first A magnetic dipole The radius of the circle corresponding to the axial coordinates of the latitude for: ;in, Let be the radius of the Fibonacci sphere; greater than or equal to 1 and less than or equal to 1 Positive integers;

[0056] Determine the first A magnetic dipole Axis coordinates for: And the first A magnetic dipole Axis coordinates for: , obtained the The predetermined positions of a magnetic dipole on a sphere.

[0057] In some embodiments of this application, the equivalent magnetic dipole model is as follows: ;in, This indicates the location of the observation point, which is determined based on the location of the carrier. Location of the observation point The measured value of the spatial magnetic field at that location, The permeability of free space, For the first The position of a magnetic dipole for arrive The relative position vector, for arrive distance, For the first The magnetic moment of a magnetic dipole The number of magnetic dipoles. This is the label for a magnetic dipole.

[0058] The magnetic field observation model is ;in, Indicates that the magnetic sensor is in Spatial magnetic field measurement at time and Spatial magnetic field measurement at time , Less than or equal to positive integers, This represents the number of magnetic sensors. express Predicted values ​​of spatial magnetic field at time and Predicted spatial magnetic field values ​​at any given time; This indicates observation noise.

[0059] In other words, the indoor magnetic field is formed by the combined action of multiple dispersed magnetic sources, mainly including electronic equipment, metal structures, pipelines, and other magnetic materials. According to classical electromagnetic theory, Helmholtz's theorem states that any static magnetic field distribution... It can be uniquely decomposed into an irrotational field (by the magnetoscale potential). (Description) and no scatter field (by current density) The superposition of descriptions is specifically represented as follows: , ;in, For Hamiltonian operators, It is a magnetic vector potential, in the region without free current. The magnetic field degenerates into an irrotational field, that is... At this point, the magnetic field distribution is determined solely by the boundary conditions and the magnetization properties of the medium, and is independent of the physical structure of the actual source.

[0060] In complex indoor magnetic environments, the physical structure and spatial distribution of real magnetic sources (such as building steel bars, electronic equipment, etc.) are difficult to model accurately. However, according to the equivalence theory, no matter how complex the real magnetic source is, there always exists a set of equivalent magnetic dipoles, whose magnetic field distribution in the observation area is consistent with the real field distribution. Therefore, this application introduces an equivalent magnetic dipole model, and by constructing virtual magnetic dipoles, the magnetic field generated by them is consistent with the real magnetic field in the observation area.

[0061] Under static magnetic field and linear isotropic medium conditions, the magnetic field satisfies the superposition principle: 1) Linear superposition: The total magnetic field generated by multiple independent magnetic sources is the vector sum of the magnetic fields of each source; 2) Spatial independence: The magnetic field generated by each magnetic source is only related to its own parameters and relative position, and is not affected by other sources.

[0062] Under these conditions, the total magnetic field in a complex magnetic environment can be decomposed into a linear superposition of the background field (such as the Earth's magnetic field) and all local magnetic sources (steel bars, equipment, etc.). By equating the background field and all local magnetic sources together with the magnetic field generated by a magnetic dipole, an equivalent magnetic dipole model can be obtained. The position and magnetic moment of the magnetic dipole can be determined by parametric inversion of the equivalent magnetic dipole model.

[0063] In some embodiments of this application, a magnetic field observation model is used. The inversion of the equivalent magnetic dipole model based on the spatial magnetic field measurements at time t is performed in... The equivalent magnetic dipole model parameters at time t may include:

[0064] Will Spatial magnetic field measurement at time As an equivalent magnetic dipole model ;

[0065] The equivalent magnetic dipole model was inverted using a nonlinear optimization algorithm to obtain the parameters of the equivalent magnetic dipole model. The equivalent magnetic dipole model parameters at time t; the equivalent magnetic dipole model parameters include at least the following: The position of the magnetic dipole at time t and The magnetic dipole moment at time t.

[0066] The problem of finding the equivalent magnetic dipole is essentially a nonlinear least squares optimization problem. The existence of its solution depends on the sufficiency of the observation data. Based on the relationship between the number of parameters and the degrees of freedom of measurement, each magnetic dipole requires 6 parameters, including the position of the magnetic dipole. and magnetic moment There are a total of An unknown variable, subscript , and Representing the coordinates of the carrier respectively axis, shaft and Axis components, subscript Indicates the first One magnetic dipole. One magnetic sensor provides three independent measurements, therefore a minimum of [number] magnetic dipoles are required. A magnetic sensor.

[0067] To reduce the number of magnetic sensors, lower positioning costs and computational complexity, and simultaneously achieve effective sampling and modeling of the spatial distribution of the magnetic field, this application employs a fixed-position spherically distributed equivalent magnetic dipole model to approximate the indoor magnetic field distribution. In the layout of the equivalent magnetic dipole positions, the Fibonacci spherical sampling method is introduced, using the golden ratio to generate uniformly distributed sampling points on the sphere, thereby placing the magnetic dipoles at these predetermined positions.

[0068] Using this method, the position of the equivalent magnetic dipole can be directly determined, and the only remaining parameter to be determined is the magnetic moment, reducing the number of parameters to be determined to... This method can ensure the uniformity of spatial coverage, given a number of dipoles. Under these conditions, the model's representational ability is maximized while avoiding overfitting problems that may result from random or heuristic layouts.

[0069] Figure 2 This is a schematic diagram of the distribution and coverage of magnetic dipoles in an equivalent magnetic dipole model based on Fibonacci spatial sampling. The blue dots represent equivalent magnetic dipoles uniformly distributed on the Fibonacci sphere after sampling. The sphere represents the Fibonacci sphere, the plane at the center of the sphere represents the magnetic field within the coverage area of ​​the magnetic array derived from the equivalent magnetic dipole inversion, and the rightmost bar represents the range of magnetic field strength values.

[0070] exist Figure 2 In the example shown, the Fibonacci spherical sampling method is used in a radius of... Sixteen equivalent magnetic dipoles are evenly arranged on the surface of the sphere, and axis, shaft and The unit for axes is meters (m). Figure 2It can be seen that the magnetic field distribution obtained by inversion in the magnetic array region exhibits good spherical uniformity and geometric symmetry. This indicates that the spatial layout scheme is complete and representative in characterizing the spatial properties of the magnetic field, and can provide a sufficient information basis for subsequent magnetic field modeling.

[0071] In some embodiments of this application, the magnetic field observation model In include Time and time The spatial magnetic field measurements were collected by each magnetic sensor. ; It is the transpose symbol;

[0072] include Time and time The measurement noise corresponding to each magnetic sensor. ;in, to for time The measurement noise corresponding to each magnetic sensor. to for time Each magnetic sensor corresponds to a measured noise.

[0073] include Time and time Each magnetic sensor obtains the predicted value of the space magnetic field using an equivalent magnetic dipole model. ;in, to The result is predicted by the equivalent magnetic dipole model. Predicted values ​​of spatial magnetic field at time. to The result is predicted by the equivalent magnetic dipole model. Predicted values ​​of the spatial magnetic field at any given time.

[0074] In some real-time methods, the parameters of the equivalent magnetic dipole model are solved and Time's up The carrier pose transformation parameters at any given time may include:

[0075] Sure Time's up The relationship between the pose transformation of the carrier at any time ,and ; for Time's up The rotation matrix at time step, As a carrier Time relative to The displacement at time t, this displacement is at Representation of the time carrier coordinate system; As a carrier Time relative to The rotation angle at any given moment; express Time-array magnetic sensors in Position of the carrier in the coordinate system at any given time express The position of the time-array magnetic sensor in the carrier coordinate system;

[0076] Determine the first based on the pose transformation relationship A magnetic dipole in Time and The relative position vector at any given moment;

[0077] Substituting the relative position vector into the equivalent magnetic dipole model, we obtain the result expressed by the parameters of the equivalent magnetic dipole model and the carrier pose transformation parameters. ;

[0078] Based on magnetic field observation model and Solution The equivalent magnetic dipole model parameters and Time's up The carrier pose transformation parameters at time t; among which, the equivalent magnetic dipole model parameters include the magnetic dipole magnetic moment. , Time's up The carrier pose transformation parameters at each moment include and .

[0079] Among them, the parameters of the equivalent magnetic dipole model and the carrier pose transformation parameters are expressed. It can be .

[0080] Furthermore, based on the magnetic field observation model, utilizing and Solution The steps may include, to and As known quantities in the magnetic field observation model, they are solved using a nonlinear optimization method. .

[0081] In other words, Time's up The pose transformation relationship of the carrier at any time is as follows: Figure 3 As shown. Among them, System Time-based carrier coordinate system System The time-carrier coordinate system, the overall shaded area represents the coverage of the equivalent magnetic dipole model, that is, the model coverage can include the pose of two adjacent time points.

[0082] refer to Figure 3 This pose transformation relationship can be represented as ,and This pose transformation relationship can be used to calculate... Time's up Time of the first A magnetic dipole in Time and Relative position vector at time to as well as to Therefore, we can conclude that .

[0083] In some embodiments of this application, the equivalent magnetic dipole model is... The equivalent magnetic dipole model parameters at time t, and Time's up The carrier pose transformation parameters at time t are obtained by inversion in the following manner:

[0084] Based on magnetic field observation model Time and Spatial magnetic field measurement value at time and Time and The spatial magnetic field measurement values ​​at time 1 are used to solve the magnetic field observation model using a nonlinear optimization method to obtain the equivalent magnetic dipole model parameters and the carrier pose transformation parameters.

[0085] Among them, the equivalent magnetic dipole model parameters include the magnetic moments of each magnetic dipole; the carrier pose transformation parameters include the carrier's position in... Time's up The displacement and rotation angle at any given moment.

[0086] Figure 4 This is a flowchart illustrating another method for parameter inversion and localization of an equivalent magnetic dipole model for indoor local space provided in an embodiment of this application. For example... Figure 4 As shown, data can first be collected using an array of magnetic sensors; simultaneously, an equivalent magnetic dipole model based on Fibonacci space sampling can be constructed, along with a magnetic field observation model containing pose variables. The collected data can be used to solve for the model parameters of each model, thereby obtaining the magnetic field parameters in the indoor magnetic field equivalent magnetic dipole model and the pose transformation parameters of adjacent frames.

[0087] The technical solution provided in this application incorporates magnetic field data collected at multiple consecutive moments into the modeling process simultaneously. It explicitly introduces carrier pose information as a parameter to be estimated into the magnetic field model. By constructing a magnetic field observation equation that includes pose variables, it uses magnetic field measurements from consecutive moments to jointly solve for the equivalent magnetic dipole model parameters and the carrier pose transformation at each moment. This method effectively solves the prediction failure and error accumulation problems caused by traditional methods relying on single-moment modeling and model coverage assumptions. It has universality and can be extended to various magnetic field modeling methods such as magnetic field polynomial models and Gaussian process models.

[0088] Furthermore, this application also proposes an equivalent magnetic dipole layout method based on Fibonacci spatial sampling. It utilizes the golden ratio to generate uniformly distributed sampling points on a sphere, places magnetic dipoles in predetermined positions, maximizes spatial coverage uniformity under a given number of dipoles, improves the model's ability to represent complex magnetic fields, and avoids overfitting problems that may be caused by random or heuristic layouts.

[0089] Meanwhile, the embodiments of this application overcome the limitations of existing magnetic field odometry, which relies on single-moment modeling and the strong assumption that "the model coverage includes the sampling point at the next moment." By simultaneously incorporating magnetic field sampling data from multiple consecutive moments into the modeling process, even in scenarios with high vehicle speeds or drastic changes in the magnetic field space, prediction failures due to insufficient model coverage can be effectively avoided. Furthermore, by using continuous moment data for joint calculation, the difference between the predicted and actual values ​​at the next moment sampling point and the cumulative effect of the single-moment model are significantly suppressed, thereby greatly improving the accuracy and continuity of pose estimation.

[0090] This application also achieves a deep integration of magnetic field modeling and pose estimation. By explicitly introducing carrier pose information as a parameter to be estimated into the magnetic field model, a magnetic field observation equation containing pose variables is constructed. The model parameters and pose transformation are jointly solved using magnetic field measurements at continuous time points. This method, which couples modeling and positioning within the same parameter inversion framework, overcomes the inherent defect of the traditional "modeling first, positioning later" process where modeling errors are propagated and amplified backward. It maintains good observability in both flat magnetic field regions and regions with local distortion, significantly enhancing the robustness of the system in complex indoor environments.

[0091] The technical solutions of this application combine universality and preferred implementation path. Its core concept can be extended to various magnetic field modeling methods such as magnetic field polynomial models and Gaussian process models, exhibiting good scalability. As a preferred solution, this application also proposes an equivalent magnetic dipole layout method based on Fibonacci spatial sampling. This method utilizes the golden ratio to generate uniformly distributed sampling points on a sphere, maximizing spatial coverage uniformity under a given number of dipoles. This not only improves the model's ability to represent complex magnetic fields but also avoids overfitting problems that may result from random or heuristic layouts, further optimizing the accuracy and stability of pose estimation.

[0092] All of the above-mentioned optional technical solutions can be combined in any way to form the optional embodiments of this application, and will not be described in detail here.

[0093] The following are embodiments of the apparatus described in this application, which can be used to execute the embodiments of the method described in this application. For details not disclosed in the apparatus embodiments of this application, please refer to the embodiments of the method described in this application.

[0094] Figure 5 This is a schematic diagram of a device for inverting and locating parameters of an equivalent magnetic dipole model for an indoor local space, provided in an embodiment of this application. Figure 5 As shown, the device includes:

[0095] Measurement module 501 is configured to set up an array of magnetic sensors in a localized indoor space to acquire data from the array of magnetic sensors. Spatial magnetic field measurement value at time and Measurement of the spatial magnetic field at a given time.

[0096] Module 502 is configured to construct an equivalent magnetic dipole model and a magnetic field observation model; the equivalent magnetic dipole model is used to predict the space magnetic field; the magnetic field observation model is used to characterize the relationship between the measured value of the space magnetic field and the predicted value of the space magnetic field.

[0097] Inversion module 503, based on the constructed equivalent magnetic dipole model and magnetic field observation model, utilizes... Time and Spatial magnetic field measurements at time t and predictions from the equivalent magnetic dipole model Time and Predicted spatial magnetic field values ​​at time t, and calculation of equivalent magnetic dipole model parameters. Time's up The carrier pose transformation parameters at any given time.

[0098] The positioning module 504 is configured to position the carrier based on the parameters of the equivalent magnetic dipole model and the carrier pose transformation parameters.

[0099] According to the technical solution provided in the embodiments of this application, magnetic field data collected at multiple consecutive moments are used simultaneously for modeling, and the carrier pose information is explicitly introduced into the magnetic field model as the parameter to be estimated. By constructing a magnetic field observation equation that includes pose variables, the equivalent magnetic dipole model parameters and the pose transformation of the carrier at each moment are jointly solved using the magnetic field measurement values ​​at consecutive moments, thereby achieving deep integration of modeling and positioning. The technical solution provided in the embodiments of this application breaks through the limitations of traditional methods that rely on single-moment modeling and coverage assumptions, effectively suppresses the accumulation of model prediction errors, and improves the accuracy and continuity of pose estimation in complex indoor magnetic field environments.

[0100] In some implementations, the positions of the magnetic dipoles in the equivalent magnetic dipole model are determined by: determining the golden ratio of each magnetic dipole: Determine the golden angle of each magnetic dipole: ; The magnetic dipole number; Magnetic dipoles are distributed around a circle centered at coordinate point (0,0,0). On a sphere with radius r, we obtain the first... A magnetic dipole Axis coordinates: ; It is a positive integer; determine the first... A magnetic dipole The radius of the latitude circle corresponding to the axis coordinates is ;in, Let be the radius of the Fibonacci sphere; greater than or equal to 1 and less than or equal to 1 A positive integer; determine the first A magnetic dipole The axis coordinates are: And the first A magnetic dipole The axis coordinates are: , obtained the The predetermined positions of a magnetic dipole on a sphere.

[0101] In some implementations, the equivalent magnetic dipole model is: ;in, This indicates the location of the observation point, which is determined based on the location of the carrier. Location of the observation point The measured value of the spatial magnetic field at that location, The permeability of free space, For the first The position of a magnetic dipole for arrive The relative position vector, for arrive distance, For the first The magnetic moment of a magnetic dipole The number of magnetic dipoles. The magnetic dipole is labeled; the magnetic field observation model is... ;in, Indicates that the magnetic sensor is in Spatial magnetic field measurement at time and Spatial magnetic field measurement at time , Less than or equal to positive integers, This represents the number of magnetic sensors. express Predicted values ​​of spatial magnetic field at time and Predicted spatial magnetic field values ​​at any given time; This indicates observation noise.

[0102] In some implementations, include Time and time The spatial magnetic field measurements were collected by each magnetic sensor. ; It is the transpose symbol; include Time and time The measurement noise corresponding to each magnetic sensor. ;in, to for time The measurement noise corresponding to each magnetic sensor. to for time The measurement noise corresponding to each magnetic sensor; include Time and time Each magnetic sensor obtains the predicted value of the space magnetic field using an equivalent magnetic dipole model. ;in, to The result is predicted by the equivalent magnetic dipole model. Predicted values ​​of spatial magnetic field at time. to The result is predicted by the equivalent magnetic dipole model. Predicted values ​​of the spatial magnetic field at any given time.

[0103] In some implementations, the equivalent magnetic dipole model parameters are solved and Time's up The carrier pose transformation parameters at each moment include: [determined] Time's up The relationship between the pose transformation of the carrier at any time ,and ; for Time's up The rotation matrix at time step, As a carrier Time relative to Displacement at time t. As a carrier Time relative to The rotation angle at any given moment; express Time-array magnetic sensors in Position of the carrier in the coordinate system at any given time to They are respectively Time from first to second A magnetic sensor in Position of the carrier in the coordinate system at any given moment; express Position of the time-array magnetic sensor in the carrier coordinate system to They are respectively Time from first to second A magnetic sensor in Position in the carrier coordinate system at any given moment; determine the first position based on pose transformation relationship. A magnetic dipole in Time and The relative position vector at time t; substituting the relative position vector into the equivalent magnetic dipole model, we obtain the expression given by the parameters of the equivalent magnetic dipole model and the carrier pose transformation parameters. Based on the magnetic field observation model, utilizing and Solution The equivalent magnetic dipole model parameters and Time's up The carrier pose transformation parameters at time t; among which, the equivalent magnetic dipole model parameters include the magnetic dipole magnetic moment. , Time's up The carrier pose transformation parameters at each moment include and .

[0104] In some implementations, the parameters expressed by the equivalent magnetic dipole model and the carrier pose transformation parameters are... for: .

[0105] In some implementations, magnetic field observation models are used... and Solution The steps include, to and As known quantities in the magnetic field observation model, they are solved using a nonlinear optimization method. .

[0106] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0107] Figure 6 This is a schematic diagram of an electronic device provided in an embodiment of this application. Figure 6 As shown, the electronic device 6 of this embodiment includes a processor 601, a memory 602, and a computer program 603 stored in the memory 602 and executable on the processor 601. When the processor 601 executes the computer program 603, it implements the steps in the various method embodiments described above. Alternatively, when the processor 601 executes the computer program 603, it implements the functions of each module / unit in the various device embodiments described above.

[0108] Electronic device 6 can be a desktop computer, laptop, handheld computer, cloud server, or other electronic device. Electronic device 6 may include, but is not limited to, processor 601 and memory 602. Those skilled in the art will understand that... Figure 6 This is merely an example of electronic device 6 and does not constitute a limitation on electronic device 6. It may include more or fewer components than shown, or different components.

[0109] The processor 601 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0110] The memory 602 can be an internal storage unit of the electronic device 6, such as a hard disk or RAM of the electronic device 6. The memory 602 can also be an external storage device of the electronic device 6, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, Flash Card, etc., equipped on the electronic device 6. The memory 602 can also include both internal and external storage units of the electronic device 6. The memory 602 is used to store computer programs and other programs and data required by the electronic device.

[0111] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0112] If an integrated module / unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program may include computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. A computer-readable medium may include: any entity or device capable of carrying computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0113] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. An indoor local space equivalent magnetic dipole model parameter inversion and positioning method, characterized in that, include: In an indoor local space, an array magnetic sensor is arranged to acquire spatial magnetic field measurement values of the array magnetic sensor at a time instant and a time instant; An equivalent magnetic dipole model and a magnetic field observation model are constructed; the equivalent magnetic dipole model is used to predict the space magnetic field; the magnetic field observation model is used to characterize the relationship between the measured value and the predicted value of the space magnetic field. Based on the constructed equivalent magnetic dipole model and the magnetic field observation model, using Time and Spatial magnetic field measurements at time t and predictions from the equivalent magnetic dipole model Time and Predicted spatial magnetic field values ​​at time t, and calculation of equivalent magnetic dipole model parameters. Time's up The carrier pose transformation parameters at any given time; The carrier is positioned based on the equivalent magnetic dipole model parameters and the carrier pose transformation parameters.

2. The method of claim 1, wherein, The position of the magnetic dipole in the equivalent magnetic dipole model is determined in the following way: Determine the golden ratio of each magnetic dipole: ; Determine the golden angle of each magnetic dipole: ; This is the magnetic dipole number; Will Magnetic dipoles are distributed around a circle centered at coordinate point (0,0,0). On a sphere with radius r, we obtain the first... A magnetic dipole Axis coordinates : ; It is a positive integer; Determine the first A magnetic dipole The radius of the circle corresponding to the axial coordinates of the latitude for: ;in, Let be the radius of the Fibonacci sphere; greater than or equal to 1 and less than or equal to 1 Positive integers; determining the axis coordinate of the magnetic dipole and the axis coordinate of the magnetic dipole , resulting in a predetermined position of the magnetic dipole on the sphere.

3. The method for parameter inversion and localization of an equivalent magnetic dipole model for indoor local space according to claim 1 or 2, characterized in that, The equivalent magnetic dipole model is as follows: ;in, This indicates the location of the observation point, which is determined based on the location of the carrier. Location of the observation point The measured value of the spatial magnetic field at that location, The permeability of free space, For the first The position of a magnetic dipole for arrive The relative position vector, for arrive distance, For the first The magnetic moment of a magnetic dipole The number of magnetic dipoles. For magnetic dipoles; The magnetic field observation model is as follows: ;in, Indicates that the magnetic sensor is in Spatial magnetic field measurement at time and Spatial magnetic field measurement at time , Less than or equal to positive integers, This represents the number of magnetic sensors. express Predicted values ​​of spatial magnetic field at time and Predicted spatial magnetic field value at any given time; This indicates observation noise.

4. The method for parameter inversion and localization of the equivalent magnetic dipole model in indoor local space according to claim 3, characterized in that, include Time and time The spatial magnetic field measurements were collected by each magnetic sensor. ; It is the transpose symbol; include Time and time The measurement noise corresponding to each magnetic sensor. ;in, to for time The measurement noise corresponding to each magnetic sensor. to for time The measurement noise corresponding to each magnetic sensor; include Time and time The spatial magnetic field values ​​were predicted by each magnetic sensor using an equivalent magnetic dipole model. ;in, to The result is predicted by the equivalent magnetic dipole model. Predicted values ​​of spatial magnetic field at any time to The result is predicted by the equivalent magnetic dipole model. Predicted values ​​of the spatial magnetic field at any given time.

5. The method for parameter inversion and localization of an equivalent magnetic dipole model in an indoor local space according to claim 4, characterized in that, Solving the equivalent magnetic dipole model parameters and Time's up The carrier pose transformation parameters at time t include: Sure Time's up The relationship between the pose transformation of the carrier at any time ,and ; for Time's up The rotation matrix at time step, As a carrier Time relative to Displacement at time 1 / 2 As a carrier Time relative to The rotation angle at any given moment; , express Time-array magnetic sensors in Position of the carrier in the coordinate system at any given time to They are respectively Time from first to second A magnetic sensor in Position of the carrier in the coordinate system at any given moment; , express Position of the time-array magnetic sensor in the carrier coordinate system to They are respectively Time from first to second A magnetic sensor in Position of the carrier in the coordinate system at any given moment; Based on the aforementioned pose transformation relationship, determine the first... A magnetic dipole in Time and The relative position vector at any given moment; Substituting the relative position vector into the equivalent magnetic dipole model yields the result expressed by the parameters of the equivalent magnetic dipole model and the carrier pose transformation parameters. ; Based on the magnetic field observation model, utilizing and Solution The equivalent magnetic dipole model parameters and Time's up The carrier pose transformation parameters at time t; among which, the equivalent magnetic dipole model parameters include the magnetic dipole magnetic moment. , Time's up The carrier pose transformation parameters at each moment include and .

6. The method for parameter inversion and localization of an equivalent magnetic dipole model in an indoor local space according to claim 5, characterized in that, The parameters expressed by the equivalent magnetic dipole model and the carrier pose transformation parameters are as follows. for: 。 7. The method for parameter inversion and localization of an equivalent magnetic dipole model in an indoor local space according to claim 5, characterized in that, Based on the magnetic field observation model, utilizing and Solution The steps include, to and As known quantities in the magnetic field observation model, they are solved using a nonlinear optimization method. .

8. A device for parameter inversion and positioning of an equivalent magnetic dipole model in an indoor local space, characterized in that, include: The measurement module is configured to place an array of magnetic sensors in a localized indoor space to acquire data from the array of magnetic sensors. Spatial magnetic field measurement value at time and The spatial magnetic field measurement value at a given time; The construction module is configured to construct an equivalent magnetic dipole model and a magnetic field observation model; the equivalent magnetic dipole model is used to predict the space magnetic field; the magnetic field observation model is used to characterize the relationship between the measured value of the space magnetic field and the predicted value of the space magnetic field. The inversion module, based on the constructed equivalent magnetic dipole model and the magnetic field observation model, utilizes... Time and Spatial magnetic field measurements at time t and predictions from the equivalent magnetic dipole model Time and Predicted spatial magnetic field values ​​at time t, and calculation of equivalent magnetic dipole model parameters. Time's up The carrier pose transformation parameters at any given time; The positioning module is configured to position the carrier based on the equivalent magnetic dipole model parameters and the carrier pose transformation parameters.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the indoor local space equivalent magnetic dipole model parameter inversion and positioning method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the indoor local space equivalent magnetic dipole model parameter inversion and positioning method as described in any one of claims 1 to 7.

Images